{smcl}
{com}{sf}{ul off}{txt}{.-}
      name:  {res}<unnamed>
       {txt}log:  {res}/Users/jblume/Documents/GitHub/coal_finance/submissions/JFE/Accepted_Paper_Submission/Code_Replication_Package/Log_Book/Main_Tables.smcl
  {txt}log type:  {res}smcl
 {txt}opened on:  {res}12 May 2025, 15:05:08

{com}. do "/var/folders/b3/396jh0bj4tg2k96f7rpnpgdw0000gq/T//SD54558.000000"
{txt}
{com}. 
. ********************************************************************************
. ********************************************************************************
. * Table 3: Determinants of Bank Exit Policy Adoption and Strength
. ********************************************************************************
. ********************************************************************************
. use "../Intermediate/policy_determinants_panel", clear
{txt}
{com}. keep if sample_big & share_gcel_pre_agg <= 1
{txt}(13,417 observations deleted)

{com}. 
. eststo clear
{txt}
{com}. 
. reghdfe ban_intensity_2030_comb_sd  loglending_pre_agg, ///
>     noabs vce(robust)
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 1 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       229
{txt}Absorbing 1 HDFE group{col 51}F({res}   1{txt},{res}    227{txt}){col 67}= {res}     69.12
{txt}{col 51}Prob > F{col 67}= {res}    0.0000
{txt}{col 51}R-squared{col 67}= {res}    0.2762
{txt}{col 51}Adj R-squared{col 67}= {res}    0.2731
{txt}{col 51}Within R-sq.{col 67}= {res}    0.2762
{txt}{col 51}Root MSE{col 67}= {res}    0.7750

{txt}{hline 19}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 20}{c |}{col 32}    Robust
{col 1}ban_intensity~b_sd{col 20}{c |} Coefficient{col 32}  std. err.{col 44}      t{col 52}   P>|t|{col 60}     [95% con{col 73}f. interval]
{hline 19}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
loglending_pre_agg {c |}{col 20}{res}{space 2} .2487751{col 32}{space 2} .0299232{col 43}{space 1}    8.31{col 52}{space 3}0.000{col 60}{space 4} .1898122{col 73}{space 3} .3077379
{txt}{space 13}_cons {c |}{col 20}{res}{space 2}-1.606423{col 32}{space 2} .2228783{col 43}{space 1}   -7.21{col 52}{space 3}0.000{col 60}{space 4}-2.045598{col 73}{space 3}-1.167249
{txt}{hline 19}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. eststo
{txt}({res}est1{txt} stored)

{com}. 
. reghdfe ban_intensity_2030_comb_sd  loglending_pre_agg share_gcel_pre_agg, ///
>     noabs vce(robust)
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 1 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       229
{txt}Absorbing 1 HDFE group{col 51}F({res}   2{txt},{res}    226{txt}){col 67}= {res}     35.44
{txt}{col 51}Prob > F{col 67}= {res}    0.0000
{txt}{col 51}R-squared{col 67}= {res}    0.2763
{txt}{col 51}Adj R-squared{col 67}= {res}    0.2699
{txt}{col 51}Within R-sq.{col 67}= {res}    0.2763
{txt}{col 51}Root MSE{col 67}= {res}    0.7767

{txt}{hline 19}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 20}{c |}{col 32}    Robust
{col 1}ban_intensity~b_sd{col 20}{c |} Coefficient{col 32}  std. err.{col 44}      t{col 52}   P>|t|{col 60}     [95% con{col 73}f. interval]
{hline 19}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
loglending_pre_agg {c |}{col 20}{res}{space 2} .2501678{col 32}{space 2} .0339316{col 43}{space 1}    7.37{col 52}{space 3}0.000{col 60}{space 4} .1833051{col 73}{space 3} .3170306
{txt}share_gcel_pre_agg {c |}{col 20}{res}{space 2} .0237195{col 32}{space 2} .1906565{col 43}{space 1}    0.12{col 52}{space 3}0.901{col 60}{space 4}-.3519723{col 73}{space 3} .3994113
{txt}{space 13}_cons {c |}{col 20}{res}{space 2} -1.62228{col 32}{space 2} .2840571{col 43}{space 1}   -5.71{col 52}{space 3}0.000{col 60}{space 4}-2.182019{col 73}{space 3}-1.062541
{txt}{hline 19}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. eststo
{txt}({res}est2{txt} stored)

{com}. 
. reghdfe ban_intensity_2030_comb_sd  loglending_pre_agg share_gcel_pre_agg  ///
>     _b_bank_year_0913 _b_firm_year_0913, ///
>     noabs vce(robust)
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 1 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       225
{txt}Absorbing 1 HDFE group{col 51}F({res}   4{txt},{res}    220{txt}){col 67}= {res}     19.11
{txt}{col 51}Prob > F{col 67}= {res}    0.0000
{txt}{col 51}R-squared{col 67}= {res}    0.2899
{txt}{col 51}Adj R-squared{col 67}= {res}    0.2770
{txt}{col 51}Within R-sq.{col 67}= {res}    0.2899
{txt}{col 51}Root MSE{col 67}= {res}    0.7781

{txt}{hline 19}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 20}{c |}{col 32}    Robust
{col 1}ban_intensity~b_sd{col 20}{c |} Coefficient{col 32}  std. err.{col 44}      t{col 52}   P>|t|{col 60}     [95% con{col 73}f. interval]
{hline 19}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
loglending_pre_agg {c |}{col 20}{res}{space 2} .2550092{col 32}{space 2} .0348677{col 43}{space 1}    7.31{col 52}{space 3}0.000{col 60}{space 4} .1862919{col 73}{space 3} .3237266
{txt}share_gcel_pre_agg {c |}{col 20}{res}{space 2}  .106332{col 32}{space 2}  .241909{col 43}{space 1}    0.44{col 52}{space 3}0.661{col 60}{space 4}-.3704236{col 73}{space 3} .5830876
{txt}{space 1}_b_bank_year_0913 {c |}{col 20}{res}{space 2} .0910965{col 32}{space 2}  .090638{col 43}{space 1}    1.01{col 52}{space 3}0.316{col 60}{space 4}-.0875334{col 73}{space 3} .2697264
{txt}{space 1}_b_firm_year_0913 {c |}{col 20}{res}{space 2} .1441941{col 32}{space 2} .1560088{col 43}{space 1}    0.92{col 52}{space 3}0.356{col 60}{space 4}-.1632689{col 73}{space 3}  .451657
{txt}{space 13}_cons {c |}{col 20}{res}{space 2}-1.694566{col 32}{space 2}  .294269{col 43}{space 1}   -5.76{col 52}{space 3}0.000{col 60}{space 4}-2.274513{col 73}{space 3}-1.114619
{txt}{hline 19}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. eststo
{txt}({res}est3{txt} stored)

{com}. 
. reghdfe ban_intensity_2030_comb_sd  loglending_pre_agg share_gcel_pre_agg  ///
>     _b_bank_year_0913 _b_firm_year_0913 indus_esg_2020 enviro_esg_2020, ///
>     noabs vce(robust)
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 1 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       172
{txt}Absorbing 1 HDFE group{col 51}F({res}   6{txt},{res}    165{txt}){col 67}= {res}     13.97
{txt}{col 51}Prob > F{col 67}= {res}    0.0000
{txt}{col 51}R-squared{col 67}= {res}    0.3636
{txt}{col 51}Adj R-squared{col 67}= {res}    0.3404
{txt}{col 51}Within R-sq.{col 67}= {res}    0.3636
{txt}{col 51}Root MSE{col 67}= {res}    0.8119

{txt}{hline 19}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 20}{c |}{col 32}    Robust
{col 1}ban_intensity~b_sd{col 20}{c |} Coefficient{col 32}  std. err.{col 44}      t{col 52}   P>|t|{col 60}     [95% con{col 73}f. interval]
{hline 19}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
loglending_pre_agg {c |}{col 20}{res}{space 2} .1875018{col 32}{space 2} .0404798{col 43}{space 1}    4.63{col 52}{space 3}0.000{col 60}{space 4} .1075767{col 73}{space 3} .2674269
{txt}share_gcel_pre_agg {c |}{col 20}{res}{space 2} .1382645{col 32}{space 2} .3379466{col 43}{space 1}    0.41{col 52}{space 3}0.683{col 60}{space 4}-.5289928{col 73}{space 3} .8055218
{txt}{space 1}_b_bank_year_0913 {c |}{col 20}{res}{space 2} .0695297{col 32}{space 2}  .116428{col 43}{space 1}    0.60{col 52}{space 3}0.551{col 60}{space 4} -.160351{col 73}{space 3} .2994105
{txt}{space 1}_b_firm_year_0913 {c |}{col 20}{res}{space 2} .1708669{col 32}{space 2}  .193452{col 43}{space 1}    0.88{col 52}{space 3}0.378{col 60}{space 4}-.2110936{col 73}{space 3} .5528273
{txt}{space 4}indus_esg_2020 {c |}{col 20}{res}{space 2} .0730074{col 32}{space 2} .0358566{col 43}{space 1}    2.04{col 52}{space 3}0.043{col 60}{space 4} .0022105{col 73}{space 3} .1438043
{txt}{space 3}enviro_esg_2020 {c |}{col 20}{res}{space 2} .0814789{col 32}{space 2} .0312625{col 43}{space 1}    2.61{col 52}{space 3}0.010{col 60}{space 4} .0197528{col 73}{space 3} .1432051
{txt}{space 13}_cons {c |}{col 20}{res}{space 2}-1.868313{col 32}{space 2} .3501073{col 43}{space 1}   -5.34{col 52}{space 3}0.000{col 60}{space 4}-2.559581{col 73}{space 3}-1.177045
{txt}{hline 19}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. eststo
{txt}({res}est4{txt} stored)

{com}. 
. reghdfe ban_intensity_2030_comb_sd  loglending_pre_agg share_gcel_pre_agg  ///
>     _b_bank_year_0913 _b_firm_year_0913 indus_esg_2020 enviro_esg_2020 ib4.nContinent, ///
>     noabs vce(robust)
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 1 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       172
{txt}Absorbing 1 HDFE group{col 51}F({res}   9{txt},{res}    162{txt}){col 67}= {res}     17.08
{txt}{col 51}Prob > F{col 67}= {res}    0.0000
{txt}{col 51}R-squared{col 67}= {res}    0.5093
{txt}{col 51}Adj R-squared{col 67}= {res}    0.4820
{txt}{col 51}Within R-sq.{col 67}= {res}    0.5093
{txt}{col 51}Root MSE{col 67}= {res}    0.7194

{txt}{hline 19}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 20}{c |}{col 32}    Robust
{col 1}ban_intensity~b_sd{col 20}{c |} Coefficient{col 32}  std. err.{col 44}      t{col 52}   P>|t|{col 60}     [95% con{col 73}f. interval]
{hline 19}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
loglending_pre_agg {c |}{col 20}{res}{space 2} .1818321{col 32}{space 2}  .042781{col 43}{space 1}    4.25{col 52}{space 3}0.000{col 60}{space 4} .0973519{col 73}{space 3} .2663123
{txt}share_gcel_pre_agg {c |}{col 20}{res}{space 2} .3706977{col 32}{space 2} .3126287{col 43}{space 1}    1.19{col 52}{space 3}0.237{col 60}{space 4}-.2466551{col 73}{space 3} .9880504
{txt}{space 1}_b_bank_year_0913 {c |}{col 20}{res}{space 2} .1359329{col 32}{space 2} .1218333{col 43}{space 1}    1.12{col 52}{space 3}0.266{col 60}{space 4}-.1046533{col 73}{space 3}  .376519
{txt}{space 1}_b_firm_year_0913 {c |}{col 20}{res}{space 2} .1789321{col 32}{space 2} .1778388{col 43}{space 1}    1.01{col 52}{space 3}0.316{col 60}{space 4} -.172249{col 73}{space 3} .5301132
{txt}{space 4}indus_esg_2020 {c |}{col 20}{res}{space 2} .0346018{col 32}{space 2} .0313971{col 43}{space 1}    1.10{col 52}{space 3}0.272{col 60}{space 4}-.0273986{col 73}{space 3} .0966022
{txt}{space 3}enviro_esg_2020 {c |}{col 20}{res}{space 2} .0623225{col 32}{space 2} .0272828{col 43}{space 1}    2.28{col 52}{space 3}0.024{col 60}{space 4} .0084467{col 73}{space 3} .1161984
{txt}{space 18} {c |}
{space 8}nContinent {c |}
{space 13}Asia  {c |}{col 20}{res}{space 2}-.0663447{col 32}{space 2} .1153762{col 43}{space 1}   -0.58{col 52}{space 3}0.566{col 60}{space 4}  -.29418{col 73}{space 3} .1614905
{txt}{space 11}Europe  {c |}{col 20}{res}{space 2} .8202766{col 32}{space 2}  .239698{col 43}{space 1}    3.42{col 52}{space 3}0.001{col 60}{space 4} .3469412{col 73}{space 3} 1.293612
{txt}{space 4}North America  {c |}{col 20}{res}{space 2}-.3896478{col 32}{space 2} .2173175{col 43}{space 1}   -1.79{col 52}{space 3}0.075{col 60}{space 4}-.8187881{col 73}{space 3} .0394924
{txt}{space 18} {c |}
{space 13}_cons {c |}{col 20}{res}{space 2}-1.650527{col 32}{space 2} .3763088{col 43}{space 1}   -4.39{col 52}{space 3}0.000{col 60}{space 4}-2.393629{col 73}{space 3}-.9074237
{txt}{hline 19}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. eststo
{txt}({res}est5{txt} stored)

{com}. 
. 
. esttab, stats(r2 N) nonotes starlevels(* .10 ** .05 *** .01) nolegend numbers label  nomtitles drop(4.nContinent)
{res}
{txt}{hline 100}
{txt}                              (1)             (2)             (3)             (4)             (5)   
{txt}{hline 100}
{txt}Bank Size           {res}        0.249***        0.250***        0.255***        0.188***        0.182***{txt}
                    {res} {ralign 12:{txt:(}8.31{txt:)}}    {ralign 12:{txt:(}7.37{txt:)}}    {ralign 12:{txt:(}7.31{txt:)}}    {ralign 12:{txt:(}4.63{txt:)}}    {ralign 12:{txt:(}4.25{txt:)}}   {txt}

{txt}Coal Share of Lend~g{res}                       0.0237           0.106           0.138           0.371   {txt}
                    {res}                 {ralign 12:{txt:(}0.12{txt:)}}    {ralign 12:{txt:(}0.44{txt:)}}    {ralign 12:{txt:(}0.41{txt:)}}    {ralign 12:{txt:(}1.19{txt:)}}   {txt}

{txt}Bank Coal Financin~h{res}                                       0.0911          0.0695           0.136   {txt}
                    {res}                                 {ralign 12:{txt:(}1.01{txt:)}}    {ralign 12:{txt:(}0.60{txt:)}}    {ralign 12:{txt:(}1.12{txt:)}}   {txt}

{txt}Coal Borrowers' Cr~h{res}                                        0.144           0.171           0.179   {txt}
                    {res}                                 {ralign 12:{txt:(}0.92{txt:)}}    {ralign 12:{txt:(}0.88{txt:)}}    {ralign 12:{txt:(}1.01{txt:)}}   {txt}

{txt}2020 indus_esg_     {res}                                                       0.0730**        0.0346   {txt}
                    {res}                                                 {ralign 12:{txt:(}2.04{txt:)}}    {ralign 12:{txt:(}1.10{txt:)}}   {txt}

{txt}2020 enviro_esg_    {res}                                                       0.0815***       0.0623** {txt}
                    {res}                                                 {ralign 12:{txt:(}2.61{txt:)}}    {ralign 12:{txt:(}2.28{txt:)}}   {txt}

{txt}Asia                {res}                                                                      -0.0663   {txt}
                    {res}                                                                 {ralign 12:{txt:(}-0.58{txt:)}}   {txt}

{txt}Europe              {res}                                                                        0.820***{txt}
                    {res}                                                                 {ralign 12:{txt:(}3.42{txt:)}}   {txt}

{txt}North America       {res}                                                                       -0.390*  {txt}
                    {res}                                                                 {ralign 12:{txt:(}-1.79{txt:)}}   {txt}

{txt}Constant            {res}       -1.606***       -1.622***       -1.695***       -1.868***       -1.651***{txt}
                    {res} {ralign 12:{txt:(}-7.21{txt:)}}    {ralign 12:{txt:(}-5.71{txt:)}}    {ralign 12:{txt:(}-5.76{txt:)}}    {ralign 12:{txt:(}-5.34{txt:)}}    {ralign 12:{txt:(}-4.39{txt:)}}   {txt}
{txt}{hline 100}
{txt}r2                  {res}        0.276           0.276           0.290           0.364           0.509   {txt}
{txt}N                   {res}          229             229             225             172             172   {txt}
{txt}{hline 100}

{com}. esttab using "../Results/Tables/Table_3.tex", replace booktabs b(%8.3f) se(%8.3f)  ///
>                 r2 nonotes starlevels(* .10 ** .05 *** .01) nolegend numbers label interaction(" $\times$ ") drop(4.nContinent) nomtitles prehead("{c -(} \def\sym#1{c -(}\ifmmode^{c -(}#1{c )-}\else\(^{c -(}#1{c )-}\)\fi{c )-}" ///
>                                                               "\begin{c -(}tabular{c )-}{c -(}l*{c -(}@M{c )-}{c -(}r{c )-}{c )-}" ///
>                                                               "\toprule"  ///
>                                                               "& \multicolumn{c -(}5{c )-}{c -(}c{c )-}{c -(}Policy Existence and Strength{c )-}\\" ///
>                                                               "\cmidrule(r{c -(}2pt{c )-}){c -(}2-6{c )-}") 
{res}{txt}(output written to {browse  `"../Results/Tables/Table_3.tex"'})

{com}. 
.                                                                                                                           
. ********************************************************************************
. ********************************************************************************
. * Table 4: Bank Financing of Coal Activity
. ********************************************************************************
. ********************************************************************************
. use "../Intermediate/bank_financing_regressions", clear
{txt}
{com}. keep if has_coded_policy==1 & year>=2006
{txt}(339,225 observations deleted)

{com}. 
. eststo clear
{txt}
{com}. 
. reghdfe log_lend active_policy, ///
>     absorb(year BankID) vce(robust)
{res}{txt}(dropped 2 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       984
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}    895{txt}){col 67}= {res}      9.15
{txt}{col 51}Prob > F{col 67}= {res}    0.0026
{txt}{col 51}R-squared{col 67}= {res}    0.8214
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8038
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0088
{txt}{col 51}Root MSE{col 67}= {res}    0.7513

{txt}{hline 14}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 15}{c |}{col 27}    Robust
{col 1}     log_lend{col 15}{c |} Coefficient{col 27}  std. err.{col 39}      t{col 47}   P>|t|{col 55}     [95% con{col 68}f. interval]
{hline 14}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
active_policy {c |}{col 15}{res}{space 2}-.2975676{col 27}{space 2} .0983812{col 38}{space 1}   -3.02{col 47}{space 3}0.003{col 55}{space 4}-.4906524{col 68}{space 3}-.1044828
{txt}{space 8}_cons {c |}{col 15}{res}{space 2} 7.207113{col 27}{space 2} .0292741{col 38}{space 1}  246.19{col 47}{space 3}0.000{col 55}{space 4} 7.149659{col 68}{space 3} 7.264567
{txt}{hline 14}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 13}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text} Absorbed FE{col 14}{c |} Categories{col 27} - Redundant{col 39}  = Num. Coefs{col 54}{c |}
{res}{col 1}{text}{hline 13}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}        year{col 14}{c |}{space 1}       16{col 27}{space 1}        0{col 39}{result}{space 1}       16{col 53}{text} {col 54}{c |}
{res}{col 1}{text}      BankID{col 14}{c |}{space 1}       73{col 27}{space 1}        1{col 39}{result}{space 1}       72{col 53}{text} {col 54}{c |}
{res}{col 1}{text}{hline 13}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est1{txt} stored)

{com}. 
. reghdfe log_lend active_policy c.post15#c.ban_intensity_new1_max_sd, ///
>     absorb(year BankID) vce(robust)
{res}{txt}(dropped 2 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       984
{txt}Absorbing 2 HDFE groups{col 51}F({res}   2{txt},{res}    894{txt}){col 67}= {res}      8.04
{txt}{col 51}Prob > F{col 67}= {res}    0.0003
{txt}{col 51}R-squared{col 67}= {res}    0.8228
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8052
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0168
{txt}{col 51}Root MSE{col 67}= {res}    0.7487

{txt}{hline 37}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 38}{c |}{col 50}    Robust
{col 1}                            log_lend{col 38}{c |} Coefficient{col 50}  std. err.{col 62}      t{col 70}   P>|t|{col 78}     [95% con{col 91}f. interval]
{hline 37}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 23}active_policy {c |}{col 38}{res}{space 2}-.2356104{col 50}{space 2} .0987488{col 61}{space 1}   -2.39{col 70}{space 3}0.017{col 78}{space 4}-.4294168{col 91}{space 3}-.0418039
{txt}{space 36} {c |}
c.post15#c.ban_intensity_new1_max_sd {c |}{col 38}{res}{space 2}-.1399125{col 50}{space 2} .0518673{col 61}{space 1}   -2.70{col 70}{space 3}0.007{col 78}{space 4}-.2417083{col 91}{space 3}-.0381167
{txt}{space 36} {c |}
{space 31}_cons {c |}{col 38}{res}{space 2} 7.281584{col 50}{space 2} .0395705{col 61}{space 1}  184.02{col 70}{space 3}0.000{col 78}{space 4} 7.203922{col 91}{space 3} 7.359246
{txt}{hline 37}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 13}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text} Absorbed FE{col 14}{c |} Categories{col 27} - Redundant{col 39}  = Num. Coefs{col 54}{c |}
{res}{col 1}{text}{hline 13}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}        year{col 14}{c |}{space 1}       16{col 27}{space 1}        0{col 39}{result}{space 1}       16{col 53}{text} {col 54}{c |}
{res}{col 1}{text}      BankID{col 14}{c |}{space 1}       73{col 27}{space 1}        1{col 39}{result}{space 1}       72{col 53}{text} {col 54}{c |}
{res}{col 1}{text}{hline 13}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est2{txt} stored)

{com}. 
. reghdfe log_lend active_policy c.post15#c.rf_score_sd, ///
>     absorb(year BankID) vce(robust)
{res}{txt}(dropped 2 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       984
{txt}Absorbing 2 HDFE groups{col 51}F({res}   2{txt},{res}    894{txt}){col 67}= {res}      9.77
{txt}{col 51}Prob > F{col 67}= {res}    0.0001
{txt}{col 51}R-squared{col 67}= {res}    0.8239
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8063
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0226
{txt}{col 51}Root MSE{col 67}= {res}    0.7465

{txt}{hline 23}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 24}{c |}{col 36}    Robust
{col 1}              log_lend{col 24}{c |} Coefficient{col 36}  std. err.{col 48}      t{col 56}   P>|t|{col 64}     [95% con{col 77}f. interval]
{hline 23}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 9}active_policy {c |}{col 24}{res}{space 2} -.222911{col 36}{space 2}  .095883{col 47}{space 1}   -2.32{col 56}{space 3}0.020{col 64}{space 4} -.411093{col 77}{space 3} -.034729
{txt}{space 22} {c |}
c.post15#c.rf_score_sd {c |}{col 24}{res}{space 2}-.1961519{col 36}{space 2}  .053851{col 47}{space 1}   -3.64{col 56}{space 3}0.000{col 64}{space 4} -.301841{col 77}{space 3}-.0904628
{txt}{space 22} {c |}
{space 17}_cons {c |}{col 24}{res}{space 2} 7.281469{col 36}{space 2} .0355454{col 47}{space 1}  204.85{col 56}{space 3}0.000{col 64}{space 4} 7.211707{col 77}{space 3} 7.351231
{txt}{hline 23}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 13}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text} Absorbed FE{col 14}{c |} Categories{col 27} - Redundant{col 39}  = Num. Coefs{col 54}{c |}
{res}{col 1}{text}{hline 13}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}        year{col 14}{c |}{space 1}       16{col 27}{space 1}        0{col 39}{result}{space 1}       16{col 53}{text} {col 54}{c |}
{res}{col 1}{text}      BankID{col 14}{c |}{space 1}       73{col 27}{space 1}        1{col 39}{result}{space 1}       72{col 53}{text} {col 54}{c |}
{res}{col 1}{text}{hline 13}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est3{txt} stored)

{com}. 
. reghdfe log_lend active_policy c.post15#c.rf_phaseout_score_sd, ///
>     absorb(year BankID) vce(robust)
{res}{txt}(dropped 1 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       899
{txt}Absorbing 2 HDFE groups{col 51}F({res}   2{txt},{res}    815{txt}){col 67}= {res}     10.92
{txt}{col 51}Prob > F{col 67}= {res}    0.0000
{txt}{col 51}R-squared{col 67}= {res}    0.8297
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8124
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0196
{txt}{col 51}Root MSE{col 67}= {res}    0.7509

{txt}{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                       log_lend{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 18}active_policy {c |}{col 33}{res}{space 2}-.2412213{col 45}{space 2} .1035214{col 56}{space 1}   -2.33{col 65}{space 3}0.020{col 73}{space 4}-.4444213{col 86}{space 3}-.0380213
{txt}{space 31} {c |}
c.post15#c.rf_phaseout_score_sd {c |}{col 33}{res}{space 2} -.164362{col 45}{space 2} .0459094{col 56}{space 1}   -3.58{col 65}{space 3}0.000{col 73}{space 4}-.2544766{col 86}{space 3}-.0742474
{txt}{space 31} {c |}
{space 26}_cons {c |}{col 33}{res}{space 2} 7.204492{col 45}{space 2}  .030972{col 56}{space 1}  232.61{col 65}{space 3}0.000{col 73}{space 4} 7.143697{col 86}{space 3} 7.265286
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 13}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text} Absorbed FE{col 14}{c |} Categories{col 27} - Redundant{col 39}  = Num. Coefs{col 54}{c |}
{res}{col 1}{text}{hline 13}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}        year{col 14}{c |}{space 1}       16{col 27}{space 1}        0{col 39}{result}{space 1}       16{col 53}{text} {col 54}{c |}
{res}{col 1}{text}      BankID{col 14}{c |}{space 1}       67{col 27}{space 1}        1{col 39}{result}{space 1}       66{col 53}{text} {col 54}{c |}
{res}{col 1}{text}{hline 13}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est4{txt} stored)

{com}. 
. reghdfe log_lend active_policy c.post15#c.complexity_score_max_sd, ///
>     absorb(year BankID) vce(robust)
{res}{txt}(dropped 2 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       984
{txt}Absorbing 2 HDFE groups{col 51}F({res}   2{txt},{res}    894{txt}){col 67}= {res}      7.99
{txt}{col 51}Prob > F{col 67}= {res}    0.0004
{txt}{col 51}R-squared{col 67}= {res}    0.8228
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8051
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0166
{txt}{col 51}Root MSE{col 67}= {res}    0.7488

{txt}{hline 35}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 36}{c |}{col 48}    Robust
{col 1}                          log_lend{col 36}{c |} Coefficient{col 48}  std. err.{col 60}      t{col 68}   P>|t|{col 76}     [95% con{col 89}f. interval]
{hline 35}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 21}active_policy {c |}{col 36}{res}{space 2}-.2562591{col 48}{space 2} .0971392{col 59}{space 1}   -2.64{col 68}{space 3}0.008{col 76}{space 4}-.4469065{col 89}{space 3}-.0656117
{txt}{space 34} {c |}
c.post15#c.complexity_score_max_sd {c |}{col 36}{res}{space 2} -.132284{col 48}{space 2} .0497924{col 59}{space 1}   -2.66{col 68}{space 3}0.008{col 76}{space 4}-.2300076{col 89}{space 3}-.0345603
{txt}{space 34} {c |}
{space 29}_cons {c |}{col 36}{res}{space 2} 7.286947{col 48}{space 2} .0415596{col 59}{space 1}  175.34{col 68}{space 3}0.000{col 76}{space 4} 7.205382{col 89}{space 3} 7.368513
{txt}{hline 35}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 13}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text} Absorbed FE{col 14}{c |} Categories{col 27} - Redundant{col 39}  = Num. Coefs{col 54}{c |}
{res}{col 1}{text}{hline 13}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}        year{col 14}{c |}{space 1}       16{col 27}{space 1}        0{col 39}{result}{space 1}       16{col 53}{text} {col 54}{c |}
{res}{col 1}{text}      BankID{col 14}{c |}{space 1}       73{col 27}{space 1}        1{col 39}{result}{space 1}       72{col 53}{text} {col 54}{c |}
{res}{col 1}{text}{hline 13}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est5{txt} stored)

{com}. 
. reghdfe log_lend active_policy_initial c.post15#c.ban_intensity_initial_max_sd, ///
>     absorb(year BankID) vce(robust)
{res}{txt}(dropped 2 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}       984
{txt}Absorbing 2 HDFE groups{col 51}F({res}   2{txt},{res}    894{txt}){col 67}= {res}      7.46
{txt}{col 51}Prob > F{col 67}= {res}    0.0006
{txt}{col 51}R-squared{col 67}= {res}    0.8226
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8049
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0154
{txt}{col 51}Root MSE{col 67}= {res}    0.7492

{txt}{hline 40}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 41}{c |}{col 53}    Robust
{col 1}                               log_lend{col 41}{c |} Coefficient{col 53}  std. err.{col 65}      t{col 73}   P>|t|{col 81}     [95% con{col 94}f. interval]
{hline 40}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 18}active_policy_initial {c |}{col 41}{res}{space 2}-.2851315{col 53}{space 2} .0995867{col 64}{space 1}   -2.86{col 73}{space 3}0.004{col 81}{space 4}-.4805824{col 94}{space 3}-.0896805
{txt}{space 39} {c |}
c.post15#c.ban_intensity_initial_max_sd {c |}{col 41}{res}{space 2} -.105949{col 53}{space 2}  .049182{col 64}{space 1}   -2.15{col 73}{space 3}0.031{col 81}{space 4}-.2024747{col 94}{space 3}-.0094232
{txt}{space 39} {c |}
{space 34}_cons {c |}{col 41}{res}{space 2} 7.264581{col 53}{space 2} .0389347{col 64}{space 1}  186.58{col 73}{space 3}0.000{col 81}{space 4} 7.188167{col 94}{space 3} 7.340995
{txt}{hline 40}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 13}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text} Absorbed FE{col 14}{c |} Categories{col 27} - Redundant{col 39}  = Num. Coefs{col 54}{c |}
{res}{col 1}{text}{hline 13}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}        year{col 14}{c |}{space 1}       16{col 27}{space 1}        0{col 39}{result}{space 1}       16{col 53}{text} {col 54}{c |}
{res}{col 1}{text}      BankID{col 14}{c |}{space 1}       73{col 27}{space 1}        1{col 39}{result}{space 1}       72{col 53}{text} {col 54}{c |}
{res}{col 1}{text}{hline 13}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est6{txt} stored)

{com}. 
. esttab, label starlevels(* .10 ** .05 *** .01) se(%8.3f) ///
>     order(active_policy active_policy_initial)
{res}
{txt}{hline 116}
{txt}                              (1)             (2)             (3)             (4)             (5)             (6)   
{txt}                         log_lend        log_lend        log_lend        log_lend        log_lend        log_lend   
{txt}{hline 116}
{txt}active_policy       {res}       -0.298***       -0.236**        -0.223**        -0.241**        -0.256***                {txt}
                    {res} {ralign 12:{txt:(}0.098{txt:)}}    {ralign 12:{txt:(}0.099{txt:)}}    {ralign 12:{txt:(}0.096{txt:)}}    {ralign 12:{txt:(}0.104{txt:)}}    {ralign 12:{txt:(}0.097{txt:)}}                   {txt}

{txt}active_policy_init~l{res}                                                                                       -0.285***{txt}
                    {res}                                                                                 {ralign 12:{txt:(}0.100{txt:)}}   {txt}

{txt}post15 # ban_inten~_{res}                       -0.140***                                                                {txt}
                    {res}                 {ralign 12:{txt:(}0.052{txt:)}}                                                                   {txt}

{txt}post15 # rf_score_sd{res}                                       -0.196***                                                {txt}
                    {res}                                 {ralign 12:{txt:(}0.054{txt:)}}                                                   {txt}

{txt}post15 # rf_phaseo~d{res}                                                       -0.164***                                {txt}
                    {res}                                                 {ralign 12:{txt:(}0.046{txt:)}}                                   {txt}

{txt}post15 # complexit~d{res}                                                                       -0.132***                {txt}
                    {res}                                                                 {ralign 12:{txt:(}0.050{txt:)}}                   {txt}

{txt}post15 # ban_inten~m{res}                                                                                       -0.106** {txt}
                    {res}                                                                                 {ralign 12:{txt:(}0.049{txt:)}}   {txt}

{txt}Constant            {res}        7.207***        7.282***        7.281***        7.204***        7.287***        7.265***{txt}
                    {res} {ralign 12:{txt:(}0.029{txt:)}}    {ralign 12:{txt:(}0.040{txt:)}}    {ralign 12:{txt:(}0.036{txt:)}}    {ralign 12:{txt:(}0.031{txt:)}}    {ralign 12:{txt:(}0.042{txt:)}}    {ralign 12:{txt:(}0.039{txt:)}}   {txt}
{txt}{hline 116}
{txt}Observations        {res}          984             984             984             899             984             984   {txt}
{txt}{hline 116}
{txt}Standard errors in parentheses
{txt}* p<.10, ** p<.05, *** p<.01

{com}. 
. 
. esttab using "../Results/Tables/Table_4.tex", replace booktabs b(%8.3f) se(%8.3f)  order(active_policy active_policy_initial) ///
>                 nocons nonotes starlevels(* .10 ** .05 *** .01) nolegend numbers ///
>                 label substitute("\_" "_") interaction(" $\times$ ") ///
>                 nomtitles prehead("{c -(} \def\sym#1{c -(}\ifmmode^{c -(}#1{c )-}\else\(^{c -(}#1{c )-}\)\fi{c )-}" ///
>                                                               "\begin{c -(}tabular{c )-}{c -(}l*{c -(}@M{c )-}{c -(}r{c )-}{c )-}" ///
>                                                               "\toprule" ///
>                                                               "& \multicolumn{c -(}6{c )-}{c -(}c{c )-}{c -(}Coal Debt Origination (log){c )-} \\" ///
>                                                               "\cmidrule{c -(}2-7{c )-}") ///
>                 stats(bank cy N r2_a , fmt( 0 0 %8.0fc 3) labels("Bank FE" "Year FE" "Observations" "Adj-R$^2$" ))
{res}{txt}(output written to {browse  `"../Results/Tables/Table_4.tex"'})

{com}. 
. ********************************************************************************
. ********************************************************************************
. * Table 5: Bank Exit Policies Effects: Isolating the Supply Channel
. ********************************************************************************
. ********************************************************************************
. use "../Intermediate/bank_supply_chain_regressions", clear
{txt}
{com}. 
. eststo clear
{txt}
{com}. 
. reghdfe log_1p_borr_fbt ban_intensity_new1, absorb(borrower_id#year BankID) cluster(BankID borrower_id)
{res}{txt}(dropped 481 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 17 iterations)
{res}{txt}Warning: VCV matrix was non-positive semi-definite; adjustment from Cameron, Gelbach & Miller applied.
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}   139,100
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}    419{txt}){col 67}= {res}      3.93
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0482
{txt}{col 51}R-squared{col 67}= {res}    0.3053
{txt}{col 51}Adj R-squared{col 67}= {res}    0.2700
{txt}{col 1}Number of clusters ({res}BankID{txt}) {col 30}= {res}     1,271{txt}{col 51}Within R-sq.{col 67}= {res}    0.0002
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       420{txt}{col 51}Root MSE{col 67}= {res}    1.7314

{txt}{ralign 84:(Std. err. adjusted for {res:420} clusters in {res:BankID borrower_id})}
{hline 19}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 20}{c |}{col 32}    Robust
{col 1}   log_1p_borr_fbt{col 20}{c |} Coefficient{col 32}  std. err.{col 44}      t{col 52}   P>|t|{col 60}     [95% con{col 73}f. interval]
{hline 19}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ban_intensity_new1 {c |}{col 20}{res}{space 2}-.2290921{col 32}{space 2} .1156317{col 43}{space 1}   -1.98{col 52}{space 3}0.048{col 60}{space 4}-.4563827{col 73}{space 3}-.0018015
{txt}{space 13}_cons {c |}{col 20}{res}{space 2} 1.024794{col 32}{space 2} .0038561{col 43}{space 1}  265.76{col 52}{space 3}0.000{col 60}{space 4} 1.017215{col 73}{space 3} 1.032374
{txt}{hline 19}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 20}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}        Absorbed FE{col 21}{c |} Categories{col 34} - Redundant{col 46}  = Num. Coefs{col 61}{c |}
{res}{col 1}{text}{hline 20}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}   borrower_id#year{col 21}{c |}{space 1}     5460{col 34}{space 1}     5460{col 46}{result}{space 1}        0{col 60}{text}*{col 61}{c |}
{res}{col 1}{text}             BankID{col 21}{c |}{space 1}     1271{col 34}{space 1}     1271{col 46}{result}{space 1}        0{col 60}{text}*{col 61}{c |}
{res}{col 1}{text}{hline 20}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est1{txt} stored)

{com}.         
. reghdfe log_1p_borr_fbt i.highshare#c.ban_intensity_new1, absorb(borrower_id#year BankID) cluster(BankID borrower_id)
{res}{txt}(dropped 364 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 16 iterations)
{res}{txt}Warning: VCV matrix was non-positive semi-definite; adjustment from Cameron, Gelbach & Miller applied.
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}   129,467
{txt}Absorbing 2 HDFE groups{col 51}F({res}   2{txt},{res}    385{txt}){col 67}= {res}      3.26
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0394
{txt}{col 51}R-squared{col 67}= {res}    0.3056
{txt}{col 51}Adj R-squared{col 67}= {res}    0.2703
{txt}{col 1}Number of clusters ({res}BankID{txt}) {col 30}= {res}     1,244{txt}{col 51}Within R-sq.{col 67}= {res}    0.0003
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       386{txt}{col 51}Root MSE{col 67}= {res}    1.7376

{txt}{ralign 96:(Std. err. adjusted for {res:386} clusters in {res:BankID borrower_id})}
{hline 31}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 32}{c |}{col 44}    Robust
{col 1}               log_1p_borr_fbt{col 32}{c |} Coefficient{col 44}  std. err.{col 56}      t{col 64}   P>|t|{col 72}     [95% con{col 85}f. interval]
{hline 31}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
highshare#c.ban_intensity_new1 {c |}
{space 2}$\text{Low Coal Share}_{f}$  {c |}{col 32}{res}{space 2}-.1555849{col 44}{space 2} .1510489{col 55}{space 1}   -1.03{col 64}{space 3}0.304{col 72}{space 4} -.452569{col 85}{space 3} .1413992
{txt}{space 1}$\text{High Coal Share}_{f}$  {c |}{col 32}{res}{space 2}-.4149894{col 44}{space 2} .1730753{col 55}{space 1}   -2.40{col 64}{space 3}0.017{col 72}{space 4}-.7552805{col 85}{space 3}-.0746982
{txt}{space 30} {c |}
{space 25}_cons {c |}{col 32}{res}{space 2} 1.031913{col 44}{space 2} .0039251{col 55}{space 1}  262.90{col 64}{space 3}0.000{col 72}{space 4} 1.024196{col 85}{space 3}  1.03963
{txt}{hline 31}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 20}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}        Absorbed FE{col 21}{c |} Categories{col 34} - Redundant{col 46}  = Num. Coefs{col 61}{c |}
{res}{col 1}{text}{hline 20}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}   borrower_id#year{col 21}{c |}{space 1}     5018{col 34}{space 1}     5018{col 46}{result}{space 1}        0{col 60}{text}*{col 61}{c |}
{res}{col 1}{text}             BankID{col 21}{c |}{space 1}     1244{col 34}{space 1}     1244{col 46}{result}{space 1}        0{col 60}{text}*{col 61}{c |}
{res}{col 1}{text}{hline 20}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est2{txt} stored)

{com}. 
. reghdfe log_1p_borr_fbt i.median_assets_mean#c.ban_intensity_new1, absorb(borrower_id#year BankID) cluster(BankID borrower_id)
{res}{txt}(dropped 208 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 15 iterations)
{res}{txt}Warning: VCV matrix was non-positive semi-definite; adjustment from Cameron, Gelbach & Miller applied.
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}   125,242
{txt}Absorbing 2 HDFE groups{col 51}F({res}   2{txt},{res}    317{txt}){col 67}= {res}      3.12
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0456
{txt}{col 51}R-squared{col 67}= {res}    0.2951
{txt}{col 51}Adj R-squared{col 67}= {res}    0.2635
{txt}{col 1}Number of clusters ({res}BankID{txt}) {col 30}= {res}     1,245{txt}{col 51}Within R-sq.{col 67}= {res}    0.0003
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       318{txt}{col 51}Root MSE{col 67}= {res}    1.7489

{txt}{ralign 105:(Std. err. adjusted for {res:318} clusters in {res:BankID borrower_id})}
{hline 40}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 41}{c |}{col 53}    Robust
{col 1}                        log_1p_borr_fbt{col 41}{c |} Coefficient{col 53}  std. err.{col 65}      t{col 73}   P>|t|{col 81}     [95% con{col 94}f. interval]
{hline 40}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
median_assets_mean#c.ban_intensity_new1 {c |}
{space 28}Small Firm  {c |}{col 41}{res}{space 2} -.224068{col 53}{space 2} .1595025{col 64}{space 1}   -1.40{col 73}{space 3}0.161{col 81}{space 4}-.5378853{col 94}{space 3} .0897492
{txt}{space 28}Large Firm  {c |}{col 41}{res}{space 2}-.2940097{col 53}{space 2} .1487921{col 64}{space 1}   -1.98{col 73}{space 3}0.049{col 81}{space 4}-.5867546{col 94}{space 3}-.0012647
{txt}{space 39} {c |}
{space 34}_cons {c |}{col 41}{res}{space 2} 1.040756{col 53}{space 2} .0040383{col 64}{space 1}  257.72{col 73}{space 3}0.000{col 81}{space 4}  1.03281{col 94}{space 3} 1.048701
{txt}{hline 40}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 20}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}        Absorbed FE{col 21}{c |} Categories{col 34} - Redundant{col 46}  = Num. Coefs{col 61}{c |}
{res}{col 1}{text}{hline 20}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}   borrower_id#year{col 21}{c |}{space 1}     4134{col 34}{space 1}     4134{col 46}{result}{space 1}        0{col 60}{text}*{col 61}{c |}
{res}{col 1}{text}             BankID{col 21}{c |}{space 1}     1245{col 34}{space 1}     1245{col 46}{result}{space 1}        0{col 60}{text}*{col 61}{c |}
{res}{col 1}{text}{hline 20}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est3{txt} stored)

{com}. 
. reghdfe log_1p_borr_fbt c.ban_intensity_new1 if coal_industry_mining == 0, absorb(borrower_id#year BankID) cluster(BankID borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 17 iterations)
{res}{txt}Warning: VCV matrix was non-positive semi-definite; adjustment from Cameron, Gelbach & Miller applied.
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}    63,934
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}    194{txt}){col 67}= {res}      2.65
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.1051
{txt}{col 51}R-squared{col 67}= {res}    0.3301
{txt}{col 51}Adj R-squared{col 67}= {res}    0.2918
{txt}{col 1}Number of clusters ({res}BankID{txt}) {col 30}= {res}       921{txt}{col 51}Within R-sq.{col 67}= {res}    0.0003
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       195{txt}{col 51}Root MSE{col 67}= {res}    1.7191

{txt}{ralign 84:(Std. err. adjusted for {res:195} clusters in {res:BankID borrower_id})}
{hline 19}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 20}{c |}{col 32}    Robust
{col 1}   log_1p_borr_fbt{col 20}{c |} Coefficient{col 32}  std. err.{col 44}      t{col 52}   P>|t|{col 60}     [95% con{col 73}f. interval]
{hline 19}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ban_intensity_new1 {c |}{col 20}{res}{space 2} -.240959{col 32}{space 2} .1479691{col 43}{space 1}   -1.63{col 52}{space 3}0.105{col 60}{space 4}-.5327937{col 73}{space 3} .0508757
{txt}{space 13}_cons {c |}{col 20}{res}{space 2} 1.034541{col 32}{space 2} .0049151{col 43}{space 1}  210.48{col 52}{space 3}0.000{col 60}{space 4} 1.024847{col 73}{space 3} 1.044235
{txt}{hline 19}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 20}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}        Absorbed FE{col 21}{c |} Categories{col 34} - Redundant{col 46}  = Num. Coefs{col 61}{c |}
{res}{col 1}{text}{hline 20}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}   borrower_id#year{col 21}{c |}{space 1}     2535{col 34}{space 1}     2535{col 46}{result}{space 1}        0{col 60}{text}*{col 61}{c |}
{res}{col 1}{text}             BankID{col 21}{c |}{space 1}      921{col 34}{space 1}      921{col 46}{result}{space 1}        0{col 60}{text}*{col 61}{c |}
{res}{col 1}{text}{hline 20}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est4{txt} stored)

{com}. 
. reghdfe log_1p_borr_fbt c.ban_intensity_new1 if coal_industry_mining == 1, absorb(borrower_id#year BankID) cluster(BankID borrower_id)
{res}{txt}(dropped 156 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 17 iterations)
{res}{txt}Warning: VCV matrix was non-positive semi-definite; adjustment from Cameron, Gelbach & Miller applied.
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}    75,166
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}    224{txt}){col 67}= {res}      2.49
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.1161
{txt}{col 51}R-squared{col 67}= {res}    0.3031
{txt}{col 51}Adj R-squared{col 67}= {res}    0.2653
{txt}{col 1}Number of clusters ({res}BankID{txt}) {col 30}= {res}       946{txt}{col 51}Within R-sq.{col 67}= {res}    0.0002
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       225{txt}{col 51}Root MSE{col 67}= {res}    1.7251

{txt}{ralign 84:(Std. err. adjusted for {res:225} clusters in {res:BankID borrower_id})}
{hline 19}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 20}{c |}{col 32}    Robust
{col 1}   log_1p_borr_fbt{col 20}{c |} Coefficient{col 32}  std. err.{col 44}      t{col 52}   P>|t|{col 60}     [95% con{col 73}f. interval]
{hline 19}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ban_intensity_new1 {c |}{col 20}{res}{space 2}-.2183894{col 32}{space 2} .1384518{col 43}{space 1}   -1.58{col 52}{space 3}0.116{col 60}{space 4}-.4912241{col 73}{space 3} .0544453
{txt}{space 13}_cons {c |}{col 20}{res}{space 2}  1.01651{col 32}{space 2} .0043585{col 43}{space 1}  233.23{col 52}{space 3}0.000{col 60}{space 4} 1.007921{col 73}{space 3} 1.025099
{txt}{hline 19}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 20}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}        Absorbed FE{col 21}{c |} Categories{col 34} - Redundant{col 46}  = Num. Coefs{col 61}{c |}
{res}{col 1}{text}{hline 20}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}   borrower_id#year{col 21}{c |}{space 1}     2925{col 34}{space 1}     2925{col 46}{result}{space 1}        0{col 60}{text}*{col 61}{c |}
{res}{col 1}{text}             BankID{col 21}{c |}{space 1}      946{col 34}{space 1}      946{col 46}{result}{space 1}        0{col 60}{text}*{col 61}{c |}
{res}{col 1}{text}{hline 20}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est5{txt} stored)

{com}. 
. esttab, label numbers starlevels(* .10 ** .05 *** .01) b(%8.3f) se(%8.3f)  nocons stats(N r2_a)
{res}
{txt}{hline 100}
{txt}                              (1)             (2)             (3)             (4)             (5)   
{txt}                     log_1p_bor~t    log_1p_bor~t    log_1p_bor~t    log_1p_bor~t    log_1p_bor~t   
{txt}{hline 100}
{txt}$\text{Exit Policy~,{res}       -0.229**                                        -0.241          -0.218   {txt}
                    {res} {ralign 12:{txt:(}0.116{txt:)}}                                    {ralign 12:{txt:(}0.148{txt:)}}    {ralign 12:{txt:(}0.138{txt:)}}   {txt}

{txt}$\text{Low Coal Sh~\{res}                       -0.156                                                   {txt}
                    {res}                 {ralign 12:{txt:(}0.151{txt:)}}                                                   {txt}

{txt}$\text{High Coal S~${res}                       -0.415**                                                 {txt}
                    {res}                 {ralign 12:{txt:(}0.173{txt:)}}                                                   {txt}

{txt}Small Firm # $\tex~ {res}                                       -0.224                                   {txt}
                    {res}                                 {ralign 12:{txt:(}0.160{txt:)}}                                   {txt}

{txt}Large Firm # $\tex~ {res}                                       -0.294**                                 {txt}
                    {res}                                 {ralign 12:{txt:(}0.149{txt:)}}                                   {txt}
{txt}{hline 100}
{txt}N                   {res}      1.4e+05         1.3e+05         1.3e+05         6.4e+04         7.5e+04   {txt}
{txt}r2_a                {res}        0.270           0.270           0.263           0.292           0.265   {txt}
{txt}{hline 100}
{txt}Standard errors in parentheses
{txt}* p<.10, ** p<.05, *** p<.01

{com}. 
. esttab using "../Results/Tables/Table_5.tex", replace booktabs b(%8.3f) se(%8.3f)  ///
>                 nocons nonotes starlevels(* .10 ** .05 *** .01) nolegend numbers ///
>                 label substitute("\_" "_") interaction(" $\times$ ") ///
>                 nomtitles prehead("{c -(} \def\sym#1{c -(}\ifmmode^{c -(}#1{c )-}\else\(^{c -(}#1{c )-}\)\fi{c )-}" ///
>                                                               "\begin{c -(}tabular{c )-}{c -(}l*{c -(}@M{c )-}{c -(}r{c )-}{c )-}" ///
>                                                               "\toprule" ///
>                                                               "& \multicolumn{c -(}5{c )-}{c -(}c{c )-}{c -(}Debt Issuance (log){c )-} \\" ///
>                                                               "\cmidrule{c -(}2-6{c )-}" ///
>                                                                                                                           "& & & & Power & Mining \\" ///
>                                                                                                                           "\cmidrule{c -(}5-6{c )-}") ///
>                 stats(bank borrower_year N r2_a , fmt( 0 0 %8.0fc 3) labels("Bank FE" "Borrower x Year FE" "Observations" "Adj-R$^2$" )) ///
>                                 postfoot("\bottomrule \\ \end{c -(}tabular{c )-}{c )-}")
{res}{txt}(output written to {browse  `"../Results/Tables/Table_5.tex"'})

{com}.                                 
. ********************************************************************************
. ********************************************************************************
. * Table 6: Effects of Bank Exit on Coal Firm Debt Issuance
. ********************************************************************************
. ********************************************************************************
. use ../Intermediate/analysis_panel_firmyear, clear
{txt}
{com}. keep if aggborr_pre_debt > 0 & Country != "xChina"
{txt}(1,898 observations deleted)

{com}. 
. eststo clear
{txt}
{com}. 
. reghdfe log_debt ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year) cl(borrower_id)
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 2 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     4,524
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}    347{txt}){col 67}= {res}      4.44
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0358
{txt}{col 51}R-squared{col 67}= {res}    0.5207
{txt}{col 51}Adj R-squared{col 67}= {res}    0.4791
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0017
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       348{txt}{col 51}Root MSE{col 67}= {res}    2.5747

{txt}{ralign 97:(Std. err. adjusted for {res:348} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                       log_debt{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.1530368{col 45}{space 2} .0726048{col 56}{space 1}   -2.11{col 65}{space 3}0.036{col 73}{space 4}-.2958377{col 86}{space 3}-.0102359
{txt}{space 26}_cons {c |}{col 33}{res}{space 2}  3.99064{col 45}{space 2} .0286881{col 56}{space 1}  139.10{col 65}{space 3}0.000{col 73}{space 4} 3.934215{col 86}{space 3} 4.047064
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 13}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text} Absorbed FE{col 14}{c |} Categories{col 27} - Redundant{col 39}  = Num. Coefs{col 54}{c |}
{res}{col 1}{text}{hline 13}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text} borrower_id{col 14}{c |}{space 1}      348{col 27}{space 1}      348{col 39}{result}{space 1}        0{col 53}{text}*{col 54}{c |}
{res}{col 1}{text}        year{col 14}{c |}{space 1}       13{col 27}{space 1}        0{col 39}{result}{space 1}       13{col 53}{text} {col 54}{c |}
{res}{col 1}{text}{hline 13}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE4

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:No}"

added macro:
                 e(sy) : "{res:No}"

{com}. eststo
{txt}({res}est1{txt} stored)

{com}. 
. reghdfe log_debt ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     4,238
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    325{txt}){col 67}= {res}      4.45
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0357
{txt}{col 51}R-squared{col 67}= {res}    0.6180
{txt}{col 51}Adj R-squared{col 67}= {res}    0.5317
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0016
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       326{txt}{col 51}Root MSE{col 67}= {res}    2.4264

{txt}{ralign 97:(Std. err. adjusted for {res:326} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                       log_debt{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.1985329{col 45}{space 2} .0941179{col 56}{space 1}   -2.11{col 65}{space 3}0.036{col 73}{space 4}-.3836902{col 86}{space 3}-.0133757
{txt}{space 26}_cons {c |}{col 33}{res}{space 2}  4.02892{col 45}{space 2}  .035054{col 56}{space 1}  114.93{col 65}{space 3}0.000{col 73}{space 4} 3.959958{col 86}{space 3} 4.097881
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      326{col 38}{space 1}      326{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       13{col 38}{space 1}        0{col 50}{result}{space 1}       13{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      390{col 38}{space 1}       13{col 50}{result}{space 1}      377{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est2{txt} stored)

{com}. 
. reghdfe log_debt c.ss_pre_all_debt_banintsty_n1_sd#i.highshare, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     3,926
{txt}Absorbing 4 HDFE groups{col 51}F({res}   2{txt},{res}    301{txt}){col 67}= {res}      4.46
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0124
{txt}{col 51}R-squared{col 67}= {res}    0.6247
{txt}{col 51}Adj R-squared{col 67}= {res}    0.5367
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0036
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       302{txt}{col 51}Root MSE{col 67}= {res}    2.4063

{txt}{ralign 109:(Std. err. adjusted for {res:302} clusters in {res:borrower_id})}
{hline 44}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 45}{c |}{col 57}    Robust
{col 1}                                   log_debt{col 45}{c |} Coefficient{col 57}  std. err.{col 69}      t{col 77}   P>|t|{col 85}     [95% con{col 98}f. interval]
{hline 44}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
highshare#c.ss_pre_all_debt_banintsty_n1_sd {c |}
{space 28}Low Coal Share  {c |}{col 45}{res}{space 2}-.0871008{col 57}{space 2} .1083424{col 68}{space 1}   -0.80{col 77}{space 3}0.422{col 85}{space 4}-.3003052{col 98}{space 3} .1261036
{txt}{space 27}High Coal Share  {c |}{col 45}{res}{space 2} -.394302{col 57}{space 2} .1387315{col 68}{space 1}   -2.84{col 77}{space 3}0.005{col 85}{space 4}-.6673085{col 98}{space 3}-.1212955
{txt}{space 43} {c |}
{space 38}_cons {c |}{col 45}{res}{space 2} 4.066975{col 57}{space 2} .0405701{col 68}{space 1}  100.25{col 77}{space 3}0.000{col 85}{space 4} 3.987138{col 98}{space 3} 4.146812
{txt}{hline 44}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      302{col 38}{space 1}      302{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       13{col 38}{space 1}        0{col 50}{result}{space 1}       13{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      377{col 38}{space 1}       13{col 50}{result}{space 1}      364{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est3{txt} stored)

{com}. 
. reghdfe log_debt c.ss_pre_all_debt_banintsty_n1_sd#i.median_assets_mean, ///
>     absorb(borrower_id i.count#i.year i.year) cl(borrower_id)
{res}{txt}(dropped 247 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 2 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     3,445
{txt}Absorbing 3 HDFE groups{col 51}F({res}   2{txt},{res}    264{txt}){col 67}= {res}      2.93
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0550
{txt}{col 51}R-squared{col 67}= {res}    0.6052
{txt}{col 51}Adj R-squared{col 67}= {res}    0.5190
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0039
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       265{txt}{col 51}Root MSE{col 67}= {res}    2.4518

{txt}{ralign 118:(Std. err. adjusted for {res:265} clusters in {res:borrower_id})}
{hline 53}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 54}{c |}{col 66}    Robust
{col 1}                                            log_debt{col 54}{c |} Coefficient{col 66}  std. err.{col 78}      t{col 86}   P>|t|{col 94}     [95% con{col 107}f. interval]
{hline 53}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
median_assets_mean#c.ss_pre_all_debt_banintsty_n1_sd {c |}
{space 41}Small Firm  {c |}{col 54}{res}{space 2}-.4039993{col 66}{space 2} .1668917{col 77}{space 1}   -2.42{col 86}{space 3}0.016{col 94}{space 4}-.7326075{col 107}{space 3}-.0753911
{txt}{space 41}Large Firm  {c |}{col 54}{res}{space 2} -.172549{col 66}{space 2}  .122526{col 77}{space 1}   -1.41{col 86}{space 3}0.160{col 94}{space 4}-.4138015{col 107}{space 3} .0687035
{txt}{space 52} {c |}
{space 47}_cons {c |}{col 54}{res}{space 2} 4.466545{col 66}{space 2} .0488444{col 77}{space 1}   91.44{col 86}{space 3}0.000{col 94}{space 4} 4.370371{col 107}{space 3} 4.562719
{txt}{hline 53}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 14}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}  Absorbed FE{col 15}{c |} Categories{col 28} - Redundant{col 40}  = Num. Coefs{col 55}{c |}
{res}{col 1}{text}{hline 14}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}  borrower_id{col 15}{c |}{space 1}      265{col 28}{space 1}      265{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}   count#year{col 15}{c |}{space 1}      351{col 28}{space 1}        0{col 40}{result}{space 1}      351{col 54}{text} {col 55}{c |}
{res}{col 1}{text}         year{col 15}{c |}{space 1}       13{col 28}{space 1}       13{col 40}{result}{space 1}        0{col 54}{text} {col 55}{c |}
{res}{col 1}{text}{hline 14}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3nosy

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:No}"

{com}. eststo
{txt}({res}est4{txt} stored)

{com}. 
. reghdfe log_debt ss_pre_all_debt_banintsty_n1_sd if coal_industry_mining==0, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 247 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     1,859
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    142{txt}){col 67}= {res}      0.43
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.5116
{txt}{col 51}R-squared{col 67}= {res}    0.6674
{txt}{col 51}Adj R-squared{col 67}= {res}    0.5595
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0004
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       143{txt}{col 51}Root MSE{col 67}= {res}    2.3434

{txt}{ralign 97:(Std. err. adjusted for {res:143} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                       log_debt{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.0903694{col 45}{space 2} .1373382{col 56}{space 1}   -0.66{col 65}{space 3}0.512{col 73}{space 4} -.361861{col 86}{space 3} .1811222
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 3.977182{col 45}{space 2} .0552035{col 56}{space 1}   72.05{col 65}{space 3}0.000{col 73}{space 4} 3.868055{col 86}{space 3} 4.086309
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      143{col 38}{space 1}      143{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       13{col 38}{space 1}        0{col 50}{result}{space 1}       13{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      247{col 38}{space 1}       13{col 50}{result}{space 1}      234{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est5{txt} stored)

{com}. 
. reghdfe log_debt ss_pre_all_debt_banintsty_n1_sd if coal_industry_mining==1, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 208 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,197
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    168{txt}){col 67}= {res}      3.73
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0552
{txt}{col 51}R-squared{col 67}= {res}    0.6353
{txt}{col 51}Adj R-squared{col 67}= {res}    0.5222
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0028
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       169{txt}{col 51}Root MSE{col 67}= {res}    2.4583

{txt}{ralign 97:(Std. err. adjusted for {res:169} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                       log_debt{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.2626117{col 45}{space 2} .1360178{col 56}{space 1}   -1.93{col 65}{space 3}0.055{col 73}{space 4} -.531136{col 86}{space 3} .0059127
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 4.079073{col 45}{space 2} .0475802{col 56}{space 1}   85.73{col 65}{space 3}0.000{col 73}{space 4} 3.985141{col 86}{space 3} 4.173005
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      169{col 38}{space 1}      169{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       13{col 38}{space 1}        0{col 50}{result}{space 1}       13{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      286{col 38}{space 1}       13{col 50}{result}{space 1}      273{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est6{txt} stored)

{com}. 
. 
. esttab, label nonumbers nocons starlevels(* .10 ** .05 *** .01) 
{res}
{txt}{hline 116}
{txt}                     Log Debt I~e    Log Debt I~e    Log Debt I~e    Log Debt I~e    Log Debt I~e    Log Debt I~e   
{txt}{hline 116}
{txt}$\text{Bank Exit E~}{res}       -0.153**        -0.199**                                       -0.0904          -0.263*  {txt}
                    {res} {ralign 12:{txt:(}-2.11{txt:)}}    {ralign 12:{txt:(}-2.11{txt:)}}                                    {ralign 12:{txt:(}-0.66{txt:)}}    {ralign 12:{txt:(}-1.93{txt:)}}   {txt}

{txt}Low Coal Share # $~i{res}                                      -0.0871                                                   {txt}
                    {res}                                 {ralign 12:{txt:(}-0.80{txt:)}}                                                   {txt}

{txt}High Coal Share # ~x{res}                                       -0.394***                                                {txt}
                    {res}                                 {ralign 12:{txt:(}-2.84{txt:)}}                                                   {txt}

{txt}Small Firm # $\tex~x{res}                                                       -0.404**                                 {txt}
                    {res}                                                 {ralign 12:{txt:(}-2.42{txt:)}}                                   {txt}

{txt}Large Firm # $\tex~x{res}                                                       -0.173                                   {txt}
                    {res}                                                 {ralign 12:{txt:(}-1.41{txt:)}}                                   {txt}
{txt}{hline 116}
{txt}Observations        {res}         4524            4238            3926            3445            1859            2197   {txt}
{txt}{hline 116}
{txt}t statistics in parentheses
{txt}* p<.10, ** p<.05, *** p<.01

{com}. 
. esttab using "../Results/Tables/Table_6.tex", replace booktabs b(%8.3f) se(%8.3f)  ///
>                 nocons nonotes starlevels(* .10 ** .05 *** .01) nolegend numbers ///
>                 label substitute("\_" "_") interaction(" $\times$ ") ///
>                 nomtitles prehead("{c -(} \def\sym#1{c -(}\ifmmode^{c -(}#1{c )-}\else\(^{c -(}#1{c )-}\)\fi{c )-}" ///
>                                                               "\begin{c -(}tabular{c )-}{c -(}l*{c -(}@M{c )-}{c -(}r{c )-}{c )-}" ///
>                                                               "\toprule" ///
>                                                               "& \multicolumn{c -(}6{c )-}{c -(}c{c )-}{c -(}Debt Issuance (log){c )-} \\" ///
>                                                               "\cmidrule{c -(}2-7{c )-}" ///
>                                                               "& & & & & Power & Mining \\"  /// 
>                                                               "\cmidrule(r{c -(}2pt{c )-}){c -(}6-7{c )-} ") ///
>                 stats(borrower y cy sy N r2_a , fmt(0 0 0 0 %8.0fc 3) labels("Borrower FE" "Year FE" "Country x Year FE" "Size x Year FE" "Observations" "Adj-R$^2$" ))
{res}{txt}(output written to {browse  `"../Results/Tables/Table_6.tex"'})

{com}. ********************************************************************************
. ********************************************************************************
. * Table 7: Adjustment Margin and Substitutions Channels
. ********************************************************************************
. ********************************************************************************
. use ../Intermediate/analysis_panel_firmyear, clear
{txt}
{com}. keep if aggborr_pre_debt>0 & Country!="xChina"
{txt}(1,898 observations deleted)

{com}. 
. eststo clear
{txt}
{com}. 
. reghdfe log_debt_has_coded_policy c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     4,238
{txt}Absorbing 3 HDFE groups{col 51}F({res}   1{txt},{res}    325{txt}){col 67}= {res}      5.53
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0193
{txt}{col 51}R-squared{col 67}= {res}    0.7037
{txt}{col 51}Adj R-squared{col 67}= {res}    0.6367
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0021
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       326{txt}{col 51}Root MSE{col 67}= {res}    1.9803

{txt}{ralign 97:(Std. err. adjusted for {res:326} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}      log_debt_has_coded_policy{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.1817521{col 45}{space 2}  .077276{col 56}{space 1}   -2.35{col 65}{space 3}0.019{col 73}{space 4}-.3337764{col 86}{space 3}-.0297278
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 2.566071{col 45}{space 2} .0287812{col 56}{space 1}   89.16{col 65}{space 3}0.000{col 73}{space 4}  2.50945{col 86}{space 3} 2.622692
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      326{col 38}{space 1}      326{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      390{col 38}{space 1}        0{col 50}{result}{space 1}      390{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text} {col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE2

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est1{txt} stored)

{com}. 
. reghdfe log_debt_nopol c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     4,238
{txt}Absorbing 3 HDFE groups{col 51}F({res}   1{txt},{res}    325{txt}){col 67}= {res}      0.60
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.4410
{txt}{col 51}R-squared{col 67}= {res}    0.5897
{txt}{col 51}Adj R-squared{col 67}= {res}    0.4970
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0001
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       326{txt}{col 51}Root MSE{col 67}= {res}    2.3257

{txt}{ralign 97:(Std. err. adjusted for {res:326} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                 log_debt_nopol{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.0564366{col 45}{space 2} .0731566{col 56}{space 1}   -0.77{col 65}{space 3}0.441{col 73}{space 4}-.2003569{col 86}{space 3} .0874837
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 3.154735{col 45}{space 2}  .027247{col 56}{space 1}  115.78{col 65}{space 3}0.000{col 73}{space 4} 3.101133{col 86}{space 3} 3.208338
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      326{col 38}{space 1}      326{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      390{col 38}{space 1}        0{col 50}{result}{space 1}      390{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text} {col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE2

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est2{txt} stored)

{com}. 
. reghdfe log_debt_rel_all_pre c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     4,238
{txt}Absorbing 3 HDFE groups{col 51}F({res}   1{txt},{res}    325{txt}){col 67}= {res}      7.11
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0081
{txt}{col 51}R-squared{col 67}= {res}    0.6463
{txt}{col 51}Adj R-squared{col 67}= {res}    0.5664
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0017
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       326{txt}{col 51}Root MSE{col 67}= {res}    2.2788

{txt}{ralign 97:(Std. err. adjusted for {res:326} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}           log_debt_rel_all_pre{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.1903188{col 45}{space 2} .0713773{col 56}{space 1}   -2.67{col 65}{space 3}0.008{col 73}{space 4}-.3307387{col 86}{space 3} -.049899
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 3.512614{col 45}{space 2} .0265843{col 56}{space 1}  132.13{col 65}{space 3}0.000{col 73}{space 4} 3.460315{col 86}{space 3} 3.564913
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      326{col 38}{space 1}      326{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      390{col 38}{space 1}        0{col 50}{result}{space 1}      390{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text} {col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE2

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est3{txt} stored)

{com}. 
. reghdfe log_debt_norel_all_pre c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     4,238
{txt}Absorbing 3 HDFE groups{col 51}F({res}   1{txt},{res}    325{txt}){col 67}= {res}      0.28
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.5950
{txt}{col 51}R-squared{col 67}= {res}    0.6510
{txt}{col 51}Adj R-squared{col 67}= {res}    0.5722
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0001
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       326{txt}{col 51}Root MSE{col 67}= {res}    1.9169

{txt}{ralign 97:(Std. err. adjusted for {res:326} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}         log_debt_norel_all_pre{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.0420347{col 45}{space 2} .0789881{col 56}{space 1}   -0.53{col 65}{space 3}0.595{col 73}{space 4}-.1974273{col 86}{space 3} .1133579
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 1.866213{col 45}{space 2} .0294189{col 56}{space 1}   63.44{col 65}{space 3}0.000{col 73}{space 4} 1.808337{col 86}{space 3} 1.924088
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      326{col 38}{space 1}      326{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      390{col 38}{space 1}        0{col 50}{result}{space 1}      390{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text} {col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE2

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est4{txt} stored)

{com}. 
. reghdfe log_debt_non_bank c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     4,238
{txt}Absorbing 3 HDFE groups{col 51}F({res}   1{txt},{res}    325{txt}){col 67}= {res}      2.84
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0932
{txt}{col 51}R-squared{col 67}= {res}    0.5010
{txt}{col 51}Adj R-squared{col 67}= {res}    0.3882
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0005
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       326{txt}{col 51}Root MSE{col 67}= {res}    1.7825

{txt}{ralign 97:(Std. err. adjusted for {res:326} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}              log_debt_non_bank{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2} .0785355{col 45}{space 2}   .04664{col 56}{space 1}    1.68{col 65}{space 3}0.093{col 73}{space 4} -.013219{col 86}{space 3}   .17029
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 1.115564{col 45}{space 2}  .017371{col 56}{space 1}   64.22{col 65}{space 3}0.000{col 73}{space 4}  1.08139{col 86}{space 3} 1.149738
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      326{col 38}{space 1}      326{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      390{col 38}{space 1}        0{col 50}{result}{space 1}      390{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text} {col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE2

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est5{txt} stored)

{com}. 
. reghdfe log_debt_banks c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     4,238
{txt}Absorbing 3 HDFE groups{col 51}F({res}   1{txt},{res}    325{txt}){col 67}= {res}      5.44
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0203
{txt}{col 51}R-squared{col 67}= {res}    0.6197
{txt}{col 51}Adj R-squared{col 67}= {res}    0.5338
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0020
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       326{txt}{col 51}Root MSE{col 67}= {res}    2.3998

{txt}{ralign 97:(Std. err. adjusted for {res:326} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                 log_debt_banks{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.2173043{col 45}{space 2} .0931513{col 56}{space 1}   -2.33{col 65}{space 3}0.020{col 73}{space 4}-.4005598{col 86}{space 3}-.0340487
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 3.943618{col 45}{space 2} .0346939{col 56}{space 1}  113.67{col 65}{space 3}0.000{col 73}{space 4} 3.875365{col 86}{space 3} 4.011871
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      326{col 38}{space 1}      326{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      390{col 38}{space 1}        0{col 50}{result}{space 1}      390{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text} {col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE2

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est6{txt} stored)

{com}. 
. reghdfe pos_debt c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     4,238
{txt}Absorbing 3 HDFE groups{col 51}F({res}   1{txt},{res}    325{txt}){col 67}= {res}      3.69
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0556
{txt}{col 51}R-squared{col 67}= {res}    0.5109
{txt}{col 51}Adj R-squared{col 67}= {res}    0.4003
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0016
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       326{txt}{col 51}Root MSE{col 67}= {res}    0.3822

{txt}{ralign 97:(Std. err. adjusted for {res:326} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                       pos_debt{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.0307191{col 45}{space 2} .0159935{col 56}{space 1}   -1.92{col 65}{space 3}0.056{col 73}{space 4} -.062183{col 86}{space 3} .0007448
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} .5916678{col 45}{space 2} .0059567{col 56}{space 1}   99.33{col 65}{space 3}0.000{col 73}{space 4} .5799491{col 86}{space 3} .6033864
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      326{col 38}{space 1}      326{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      390{col 38}{space 1}        0{col 50}{result}{space 1}      390{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text} {col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE2

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est7{txt} stored)

{com}. 
. reghdfe logim_debt c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 227 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 12 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,369
{txt}Absorbing 3 HDFE groups{col 51}F({res}   1{txt},{res}    296{txt}){col 67}= {res}      0.62
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.4322
{txt}{col 51}R-squared{col 67}= {res}    0.7116
{txt}{col 51}Adj R-squared{col 67}= {res}    0.6075
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0005
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       297{txt}{col 51}Root MSE{col 67}= {res}    0.9257

{txt}{ralign 97:(Std. err. adjusted for {res:297} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                     logim_debt{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.0585172{col 45}{space 2} .0743965{col 56}{space 1}   -0.79{col 65}{space 3}0.432{col 73}{space 4}-.2049303{col 86}{space 3} .0878958
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 6.855503{col 45}{space 2} .0259023{col 56}{space 1}  264.67{col 65}{space 3}0.000{col 73}{space 4} 6.804527{col 86}{space 3} 6.906479
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      297{col 38}{space 1}      297{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      266{col 38}{space 1}        0{col 50}{result}{space 1}      266{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text} {col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE2

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est8{txt} stored)

{com}. 
. reghdfe log_equity_public c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 273 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     4,238
{txt}Absorbing 3 HDFE groups{col 51}F({res}   1{txt},{res}    325{txt}){col 67}= {res}      1.03
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.3106
{txt}{col 51}R-squared{col 67}= {res}    0.3366
{txt}{col 51}Adj R-squared{col 67}= {res}    0.1867
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0001
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       326{txt}{col 51}Root MSE{col 67}= {res}    1.6264

{txt}{ralign 97:(Std. err. adjusted for {res:326} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}              log_equity_public{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2} .0397296{col 45}{space 2} .0391212{col 56}{space 1}    1.02{col 65}{space 3}0.311{col 73}{space 4}-.0372332{col 86}{space 3} .1166923
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} .5590861{col 45}{space 2} .0145706{col 56}{space 1}   38.37{col 65}{space 3}0.000{col 73}{space 4} .5304216{col 86}{space 3} .5877507
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      326{col 38}{space 1}      326{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      390{col 38}{space 1}        0{col 50}{result}{space 1}      390{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       78{col 38}{space 1}       13{col 50}{result}{space 1}       65{col 64}{text} {col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE2

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est9{txt} stored)

{com}. 
. esttab, label nocons nonotes starlevels(* .10 ** .05 *** .01) ///
>     nolegend nonumber interaction(" $\times$ ") ///
>     mgroups(A B C D, pattern(1 0 1 0 1 0 1 ))
{res}
{txt}{hline 164}
{txt}                                A                               B                               C                               D                                   
{txt}                     log_debt_h~y    log_~t_nopol    log_debt_r..    log_debt_n..    log_debt_n~k    log_debt_b~s        pos_debt      logim_debt    log_equity~c   
{txt}{hline 164}
{txt}$\text{Bank Exit E~}{res}       -0.182**       -0.0564          -0.190***      -0.0420          0.0785*         -0.217**       -0.0307*        -0.0585          0.0397   {txt}
                    {res} {ralign 12:{txt:(}-2.35{txt:)}}    {ralign 12:{txt:(}-0.77{txt:)}}    {ralign 12:{txt:(}-2.67{txt:)}}    {ralign 12:{txt:(}-0.53{txt:)}}    {ralign 12:{txt:(}1.68{txt:)}}    {ralign 12:{txt:(}-2.33{txt:)}}    {ralign 12:{txt:(}-1.92{txt:)}}    {ralign 12:{txt:(}-0.79{txt:)}}    {ralign 12:{txt:(}1.02{txt:)}}   {txt}
{txt}{hline 164}
{txt}Observations        {res}         4238            4238            4238            4238            4238            4238            4238            2369            4238   {txt}
{txt}{hline 164}

{com}. 
. esttab using "../Results/Tables/Table_7.tex", replace booktabs b(%8.3f) se(%8.3f)  ///
>                 nocons nonotes starlevels(* .10 ** .05 *** .01) nolegend nonumbers ///
>                 label substitute("\_" "_") interaction(" $\times$ ") ///
>                 nomtitles prehead("{c -(} \def\sym#1{c -(}\ifmmode^{c -(}#1{c )-}\else\(^{c -(}#1{c )-}\)\fi{c )-}" ///
>                                                               "\begin{c -(}tabularx{c )-}{c -(}1.35\textwidth{c )-}{c -(}Xl*{c -(}@M{c )-}{c -(}r{c )-}{c )-}" ///
>                                                               "\toprule" ///
>                                                               "& \multicolumn{c -(}8{c )-}{c -(}c{c )-}{c -(}Debt Issuance (log){c )-} & Equity \\" ///
>                                                               "\cmidrule{c -(}2-10{c )-}" ///
>                                                               "& \multicolumn{c -(}2{c )-}{c -(}c{c )-}{c -(}Coal Policy Bank{c )-} & \multicolumn{c -(}2{c )-}{c -(}c{c )-}{c -(}Relationship Bank{c )-}  & \multicolumn{c -(}2{c )-}{c -(}c{c )-}{c -(}Non-Bank{c )-} &                                 \multicolumn{c -(}2{c )-}{c -(}c{c )-}{c -(}Margin{c )-} \\" ///
>                                                               "\cmidrule(r{c -(}2pt{c )-}){c -(}2-3{c )-}  \cmidrule(l{c -(}2pt{c )-}){c -(}4-5{c )-} \cmidrule(l{c -(}2pt{c )-}){c -(}6-7{c )-} \cmidrule(l{c -(}2pt{c )-}){c -(}8-9{c )-} " ///
>                                                               "& Yes & No & Yes & No & Yes & No & Extensive & Intensive & \\")  ///
>                                 postfoot("\bottomrule \end{c -(}tabularx{c )-}{c )-}") ///
>                 stats(borrower cy sy N r2_a , fmt( 0 0 0 %8.0fc 3) labels("Borrower FE" "Country x Year FE" "Size x Year FE" "Observations" "Adj-R$^2$" ))
{res}{txt}(output written to {browse  `"../Results/Tables/Table_7.tex"'})

{com}. 
. ********************************************************************************
. ********************************************************************************
. * Table 8: Balance Sheet Effects
. ********************************************************************************
. ********************************************************************************
. use ../Intermediate/analysis_panel_firmyear, clear
{txt}
{com}. keep if Country!="xChina"
{txt}(0 observations deleted)

{com}. 
. ***Panel A: Long Term Debt 
. eststo clear
{txt}
{com}. 
. reghdfe log_ltd ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year) cl(borrower_id)
{res}{txt}(dropped 6 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,530
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}    327{txt}){col 67}= {res}      9.40
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0023
{txt}{col 51}R-squared{col 67}= {res}    0.8976
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8818
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0124
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       328{txt}{col 51}Root MSE{col 67}= {res}    0.7943

{txt}{ralign 97:(Std. err. adjusted for {res:328} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                        log_ltd{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.1622044{col 45}{space 2} .0528971{col 56}{space 1}   -3.07{col 65}{space 3}0.002{col 73}{space 4}-.2662658{col 86}{space 3}-.0581429
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 7.214202{col 45}{space 2} .0201737{col 56}{space 1}  357.60{col 65}{space 3}0.000{col 73}{space 4} 7.174515{col 86}{space 3} 7.253888
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 13}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text} Absorbed FE{col 14}{c |} Categories{col 27} - Redundant{col 39}  = Num. Coefs{col 54}{c |}
{res}{col 1}{text}{hline 13}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text} borrower_id{col 14}{c |}{space 1}      328{col 27}{space 1}      328{col 39}{result}{space 1}        0{col 53}{text}*{col 54}{c |}
{res}{col 1}{text}        year{col 14}{c |}{space 1}       10{col 27}{space 1}        0{col 39}{result}{space 1}       10{col 53}{text} {col 54}{c |}
{res}{col 1}{text}{hline 13}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE4

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:No}"

added macro:
                 e(sy) : "{res:No}"

{com}. eststo
{txt}({res}est1{txt} stored)

{com}. 
. reghdfe log_ltd ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 170 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 13 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,366
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    305{txt}){col 67}= {res}     11.09
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0010
{txt}{col 51}R-squared{col 67}= {res}    0.9163
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8869
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0233
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       306{txt}{col 51}Root MSE{col 67}= {res}    0.7909

{txt}{ralign 97:(Std. err. adjusted for {res:306} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                        log_ltd{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.3264541{col 45}{space 2} .0980418{col 56}{space 1}   -3.33{col 65}{space 3}0.001{col 73}{space 4} -.519378{col 86}{space 3}-.1335302
{txt}{space 26}_cons {c |}{col 33}{res}{space 2}  7.30461{col 45}{space 2} .0352029{col 56}{space 1}  207.50{col 65}{space 3}0.000{col 73}{space 4} 7.235338{col 86}{space 3} 7.373881
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      306{col 38}{space 1}      306{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      260{col 38}{space 1}       10{col 50}{result}{space 1}      250{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       60{col 38}{space 1}       10{col 50}{result}{space 1}       50{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est2{txt} stored)

{com}. 
. reghdfe log_ltd i.highshare#c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 164 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 12 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,173
{txt}Absorbing 4 HDFE groups{col 51}F({res}   2{txt},{res}    280{txt}){col 67}= {res}      5.49
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0046
{txt}{col 51}R-squared{col 67}= {res}    0.9190
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8889
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0266
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       281{txt}{col 51}Root MSE{col 67}= {res}    0.7806

{txt}{ralign 109:(Std. err. adjusted for {res:281} clusters in {res:borrower_id})}
{hline 44}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 45}{c |}{col 57}    Robust
{col 1}                                    log_ltd{col 45}{c |} Coefficient{col 57}  std. err.{col 69}      t{col 77}   P>|t|{col 85}     [95% con{col 98}f. interval]
{hline 44}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
highshare#c.ss_pre_all_debt_banintsty_n1_sd {c |}
{space 28}Low Coal Share  {c |}{col 45}{res}{space 2}-.3683456{col 57}{space 2} .1112273{col 68}{space 1}   -3.31{col 77}{space 3}0.001{col 85}{space 4}-.5872935{col 98}{space 3}-.1493978
{txt}{space 27}High Coal Share  {c |}{col 45}{res}{space 2}-.3389302{col 57}{space 2} .1310009{col 68}{space 1}   -2.59{col 77}{space 3}0.010{col 85}{space 4}-.5968018{col 98}{space 3}-.0810585
{txt}{space 43} {c |}
{space 38}_cons {c |}{col 45}{res}{space 2} 7.359732{col 57}{space 2} .0404185{col 68}{space 1}  182.09{col 77}{space 3}0.000{col 85}{space 4} 7.280169{col 98}{space 3} 7.439295
{txt}{hline 44}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      281{col 38}{space 1}      281{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      255{col 38}{space 1}       10{col 50}{result}{space 1}      245{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       60{col 38}{space 1}       10{col 50}{result}{space 1}       50{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est3{txt} stored)

{com}. 
. reghdfe log_ltd c.ss_pre_all_debt_banintsty_n1_sd#i.median_assets_mean, ///
>     absorb(borrower_id i.count#i.year i.year) cl(borrower_id)
{res}{txt}(dropped 170 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 10 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,366
{txt}Absorbing 3 HDFE groups{col 51}F({res}   2{txt},{res}    305{txt}){col 67}= {res}      8.34
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0003
{txt}{col 51}R-squared{col 67}= {res}    0.9139
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8867
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0307
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       306{txt}{col 51}Root MSE{col 67}= {res}    0.7914

{txt}{ralign 118:(Std. err. adjusted for {res:306} clusters in {res:borrower_id})}
{hline 53}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 54}{c |}{col 66}    Robust
{col 1}                                             log_ltd{col 54}{c |} Coefficient{col 66}  std. err.{col 78}      t{col 86}   P>|t|{col 94}     [95% con{col 107}f. interval]
{hline 53}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
median_assets_mean#c.ss_pre_all_debt_banintsty_n1_sd {c |}
{space 41}Small Firm  {c |}{col 54}{res}{space 2}-.3893041{col 66}{space 2} .1230689{col 77}{space 1}   -3.16{col 86}{space 3}0.002{col 94}{space 4}-.6314757{col 107}{space 3}-.1471325
{txt}{space 41}Large Firm  {c |}{col 54}{res}{space 2}-.2213772{col 66}{space 2} .0569789{col 77}{space 1}   -3.89{col 86}{space 3}0.000{col 94}{space 4}-.3334986{col 107}{space 3}-.1092558
{txt}{space 52} {c |}
{space 47}_cons {c |}{col 54}{res}{space 2} 7.286551{col 66}{space 2} .0247675{col 77}{space 1}  294.20{col 86}{space 3}0.000{col 94}{space 4} 7.237814{col 107}{space 3} 7.335287
{txt}{hline 53}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 14}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}  Absorbed FE{col 15}{c |} Categories{col 28} - Redundant{col 40}  = Num. Coefs{col 55}{c |}
{res}{col 1}{text}{hline 14}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}  borrower_id{col 15}{c |}{space 1}      306{col 28}{space 1}      306{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}   count#year{col 15}{c |}{space 1}      260{col 28}{space 1}        0{col 40}{result}{space 1}      260{col 54}{text} {col 55}{c |}
{res}{col 1}{text}         year{col 15}{c |}{space 1}       10{col 28}{space 1}       10{col 40}{result}{space 1}        0{col 54}{text} {col 55}{c |}
{res}{col 1}{text}{hline 14}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3nosy

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:No}"

{com}. eststo
{txt}({res}est4{txt} stored)

{com}. 
. reghdfe log_ltd ss_pre_all_debt_banintsty_n1_sd if coal_industry_mining==0, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 122 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 12 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     1,070
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    137{txt}){col 67}= {res}      4.62
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0334
{txt}{col 51}R-squared{col 67}= {res}    0.9231
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8845
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0307
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       138{txt}{col 51}Root MSE{col 67}= {res}    0.7327

{txt}{ralign 97:(Std. err. adjusted for {res:138} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                        log_ltd{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.3454639{col 45}{space 2} .1607928{col 56}{space 1}   -2.15{col 65}{space 3}0.033{col 73}{space 4}-.6634206{col 86}{space 3}-.0275072
{txt}{space 26}_cons {c |}{col 33}{res}{space 2}  7.64129{col 45}{space 2} .0667071{col 56}{space 1}  114.55{col 65}{space 3}0.000{col 73}{space 4} 7.509381{col 86}{space 3} 7.773199
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      138{col 38}{space 1}      138{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      170{col 38}{space 1}       10{col 50}{result}{space 1}      160{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       59{col 38}{space 1}       10{col 50}{result}{space 1}       49{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est5{txt} stored)

{com}. 
. reghdfe log_ltd ss_pre_all_debt_banintsty_n1_sd if coal_industry_mining==1, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 155 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 13 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     1,189
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    155{txt}){col 67}= {res}      5.92
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0161
{txt}{col 51}R-squared{col 67}= {res}    0.9215
{txt}{col 51}Adj R-squared{col 67}= {res}    0.8837
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0139
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       156{txt}{col 51}Root MSE{col 67}= {res}    0.8465

{txt}{ralign 97:(Std. err. adjusted for {res:156} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                        log_ltd{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.2877498{col 45}{space 2} .1182579{col 56}{space 1}   -2.43{col 65}{space 3}0.016{col 73}{space 4}-.5213549{col 86}{space 3}-.0541448
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 6.964558{col 45}{space 2} .0364833{col 56}{space 1}  190.90{col 65}{space 3}0.000{col 73}{space 4} 6.892489{col 86}{space 3} 7.036626
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      156{col 38}{space 1}      156{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      181{col 38}{space 1}       10{col 50}{result}{space 1}      171{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       59{col 38}{space 1}       10{col 50}{result}{space 1}       49{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est6{txt} stored)

{com}. 
. esttab, label starlevels(* .10 ** .05 *** .01) nocons
{res}
{txt}{hline 116}
{txt}                              (1)             (2)             (3)             (4)             (5)             (6)   
{txt}                          log_ltd         log_ltd         log_ltd         log_ltd         log_ltd         log_ltd   
{txt}{hline 116}
{txt}$\text{Bank Exit E~}{res}       -0.162***       -0.326***                                       -0.345**        -0.288** {txt}
                    {res} {ralign 12:{txt:(}-3.07{txt:)}}    {ralign 12:{txt:(}-3.33{txt:)}}                                    {ralign 12:{txt:(}-2.15{txt:)}}    {ralign 12:{txt:(}-2.43{txt:)}}   {txt}

{txt}Low Coal Share # $~i{res}                                       -0.368***                                                {txt}
                    {res}                                 {ralign 12:{txt:(}-3.31{txt:)}}                                                   {txt}

{txt}High Coal Share # ~x{res}                                       -0.339**                                                 {txt}
                    {res}                                 {ralign 12:{txt:(}-2.59{txt:)}}                                                   {txt}

{txt}Small Firm # $\tex~x{res}                                                       -0.389***                                {txt}
                    {res}                                                 {ralign 12:{txt:(}-3.16{txt:)}}                                   {txt}

{txt}Large Firm # $\tex~x{res}                                                       -0.221***                                {txt}
                    {res}                                                 {ralign 12:{txt:(}-3.89{txt:)}}                                   {txt}
{txt}{hline 116}
{txt}Observations        {res}         2530            2366            2173            2366            1070            1189   {txt}
{txt}{hline 116}
{txt}t statistics in parentheses
{txt}* p<.10, ** p<.05, *** p<.01

{com}. 
. esttab using "../Results/Tables/Table_8_Panel_A.tex", replace booktabs b(%8.3f) se(%8.3f)  ///
>                 nocons nonotes starlevels(* .10 ** .05 *** .01) nolegend numbers ///
>                 label interaction(" $\times$ ") ///
>                                 substitute("\_" "_") ///
>                 nomtitles prehead("{c -(} \def\sym#1{c -(}\ifmmode^{c -(}#1{c )-}\else\(^{c -(}#1{c )-}\)\fi{c )-}" ///
>                                                               "\begin{c -(}tabular{c )-}{c -(}l*{c -(}@M{c )-}{c -(}r{c )-}{c )-}" ///
>                                                               "\toprule" ///
>                                                               "&  \multicolumn{c -(}6{c )-}{c -(}c{c )-}{c -(}\textbf{c -(}Long-Term Debt (log){c )-}{c )-}  \\" ///
>                                                                                                                           "\cmidrule{c -(}2-7{c )-} & & & & & Power & Mining \\" ///
>                                                               "\cmidrule(l{c -(}2pt{c )-}){c -(}6-7{c )-}") ///
>                 stats(borrower y cy sy N r2_a , fmt(0 0 0 0 %8.0fc 3) labels("Borrower FE" "Year FE" "Country x Year FE" "Size x Year FE" "Observations" "Adj-R$^2$" ))
{res}{txt}(output written to {browse  `"../Results/Tables/Table_8_Panel_A.tex"'})

{com}.               
. 
. ***Panel B: Leverage  
. eststo clear
{txt}
{com}. 
. reghdfe log_lev ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year) cl(borrower_id)
{res}{txt}(dropped 6 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,530
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}    327{txt}){col 67}= {res}      4.68
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0312
{txt}{col 51}R-squared{col 67}= {res}    0.6572
{txt}{col 51}Adj R-squared{col 67}= {res}    0.6043
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0056
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       328{txt}{col 51}Root MSE{col 67}= {res}    0.6874

{txt}{ralign 97:(Std. err. adjusted for {res:328} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                        log_lev{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.0941396{col 45}{space 2} .0435146{col 56}{space 1}   -2.16{col 65}{space 3}0.031{col 73}{space 4}-.1797436{col 86}{space 3}-.0085357
{txt}{space 26}_cons {c |}{col 33}{res}{space 2}-1.495103{col 45}{space 2} .0165955{col 56}{space 1}  -90.09{col 65}{space 3}0.000{col 73}{space 4} -1.52775{col 86}{space 3}-1.462456
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 13}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text} Absorbed FE{col 14}{c |} Categories{col 27} - Redundant{col 39}  = Num. Coefs{col 54}{c |}
{res}{col 1}{text}{hline 13}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text} borrower_id{col 14}{c |}{space 1}      328{col 27}{space 1}      328{col 39}{result}{space 1}        0{col 53}{text}*{col 54}{c |}
{res}{col 1}{text}        year{col 14}{c |}{space 1}       10{col 27}{space 1}        0{col 39}{result}{space 1}       10{col 53}{text} {col 54}{c |}
{res}{col 1}{text}{hline 13}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE4

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:No}"

added macro:
                 e(sy) : "{res:No}"

{com}. eststo
{txt}({res}est1{txt} stored)

{com}. 
. reghdfe log_lev ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 170 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 13 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,366
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    305{txt}){col 67}= {res}      9.16
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0027
{txt}{col 51}R-squared{col 67}= {res}    0.7192
{txt}{col 51}Adj R-squared{col 67}= {res}    0.6203
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0180
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       306{txt}{col 51}Root MSE{col 67}= {res}    0.6821

{txt}{ralign 97:(Std. err. adjusted for {res:306} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                        log_lev{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2} -.246803{col 45}{space 2} .0815518{col 56}{space 1}   -3.03{col 65}{space 3}0.003{col 73}{space 4}-.4072783{col 86}{space 3}-.0863276
{txt}{space 26}_cons {c |}{col 33}{res}{space 2}-1.442898{col 45}{space 2}  .029282{col 56}{space 1}  -49.28{col 65}{space 3}0.000{col 73}{space 4}-1.500519{col 86}{space 3}-1.385278
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      306{col 38}{space 1}      306{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      260{col 38}{space 1}       10{col 50}{result}{space 1}      250{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       60{col 38}{space 1}       10{col 50}{result}{space 1}       50{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est2{txt} stored)

{com}. 
. reghdfe log_lev i.highshare#c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 164 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 12 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,173
{txt}Absorbing 4 HDFE groups{col 51}F({res}   2{txt},{res}    280{txt}){col 67}= {res}      4.70
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0098
{txt}{col 51}R-squared{col 67}= {res}    0.7312
{txt}{col 51}Adj R-squared{col 67}= {res}    0.6316
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0204
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       281{txt}{col 51}Root MSE{col 67}= {res}    0.6768

{txt}{ralign 109:(Std. err. adjusted for {res:281} clusters in {res:borrower_id})}
{hline 44}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 45}{c |}{col 57}    Robust
{col 1}                                    log_lev{col 45}{c |} Coefficient{col 57}  std. err.{col 69}      t{col 77}   P>|t|{col 85}     [95% con{col 98}f. interval]
{hline 44}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
highshare#c.ss_pre_all_debt_banintsty_n1_sd {c |}
{space 28}Low Coal Share  {c |}{col 45}{res}{space 2} -.286596{col 57}{space 2} .0941034{col 68}{space 1}   -3.05{col 77}{space 3}0.003{col 85}{space 4} -.471836{col 98}{space 3} -.101356
{txt}{space 27}High Coal Share  {c |}{col 45}{res}{space 2}-.2207797{col 57}{space 2} .1031868{col 68}{space 1}   -2.14{col 77}{space 3}0.033{col 85}{space 4}   -.4239{col 98}{space 3}-.0176594
{txt}{space 43} {c |}
{space 38}_cons {c |}{col 45}{res}{space 2}-1.427214{col 57}{space 2} .0333253{col 68}{space 1}  -42.83{col 77}{space 3}0.000{col 85}{space 4}-1.492814{col 98}{space 3}-1.361614
{txt}{hline 44}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      281{col 38}{space 1}      281{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      255{col 38}{space 1}       10{col 50}{result}{space 1}      245{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       60{col 38}{space 1}       10{col 50}{result}{space 1}       50{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est3{txt} stored)

{com}. 
. reghdfe log_lev c.ss_pre_all_debt_banintsty_n1_sd#i.median_assets_mean, ///
>     absorb(borrower_id i.count#i.year i.year) cl(borrower_id)
{res}{txt}(dropped 170 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 10 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,366
{txt}Absorbing 3 HDFE groups{col 51}F({res}   2{txt},{res}    305{txt}){col 67}= {res}      5.11
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0066
{txt}{col 51}R-squared{col 67}= {res}    0.7109
{txt}{col 51}Adj R-squared{col 67}= {res}    0.6197
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0216
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       306{txt}{col 51}Root MSE{col 67}= {res}    0.6827

{txt}{ralign 118:(Std. err. adjusted for {res:306} clusters in {res:borrower_id})}
{hline 53}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 54}{c |}{col 66}    Robust
{col 1}                                             log_lev{col 54}{c |} Coefficient{col 66}  std. err.{col 78}      t{col 86}   P>|t|{col 94}     [95% con{col 107}f. interval]
{hline 53}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
median_assets_mean#c.ss_pre_all_debt_banintsty_n1_sd {c |}
{space 41}Small Firm  {c |}{col 54}{res}{space 2}-.2854112{col 66}{space 2} .1011471{col 77}{space 1}   -2.82{col 86}{space 3}0.005{col 94}{space 4}-.4844456{col 107}{space 3}-.0863767
{txt}{space 41}Large Firm  {c |}{col 54}{res}{space 2} -.142034{col 66}{space 2} .0476653{col 77}{space 1}   -2.98{col 86}{space 3}0.003{col 94}{space 4}-.2358285{col 107}{space 3}-.0482395
{txt}{space 52} {c |}
{space 47}_cons {c |}{col 54}{res}{space 2}-1.463722{col 66}{space 2} .0212515{col 77}{space 1}  -68.88{col 86}{space 3}0.000{col 94}{space 4}-1.505541{col 107}{space 3}-1.421904
{txt}{hline 53}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 14}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}  Absorbed FE{col 15}{c |} Categories{col 28} - Redundant{col 40}  = Num. Coefs{col 55}{c |}
{res}{col 1}{text}{hline 14}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}  borrower_id{col 15}{c |}{space 1}      306{col 28}{space 1}      306{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}   count#year{col 15}{c |}{space 1}      260{col 28}{space 1}        0{col 40}{result}{space 1}      260{col 54}{text} {col 55}{c |}
{res}{col 1}{text}         year{col 15}{c |}{space 1}       10{col 28}{space 1}       10{col 40}{result}{space 1}        0{col 54}{text} {col 55}{c |}
{res}{col 1}{text}{hline 14}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3nosy

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:No}"

{com}. eststo
{txt}({res}est4{txt} stored)

{com}. 
. reghdfe log_lev ss_pre_all_debt_banintsty_n1_sd if coal_industry_mining==0, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 122 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 12 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     1,070
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    137{txt}){col 67}= {res}      4.14
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0438
{txt}{col 51}R-squared{col 67}= {res}    0.7574
{txt}{col 51}Adj R-squared{col 67}= {res}    0.6358
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0292
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       138{txt}{col 51}Root MSE{col 67}= {res}    0.5967

{txt}{ralign 97:(Std. err. adjusted for {res:138} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                        log_lev{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.2742543{col 45}{space 2} .1347732{col 56}{space 1}   -2.03{col 65}{space 3}0.044{col 73}{space 4} -.540759{col 86}{space 3}-.0077495
{txt}{space 26}_cons {c |}{col 33}{res}{space 2}-1.313647{col 45}{space 2} .0559125{col 56}{space 1}  -23.49{col 65}{space 3}0.000{col 73}{space 4} -1.42421{col 86}{space 3}-1.203084
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      138{col 38}{space 1}      138{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      170{col 38}{space 1}       10{col 50}{result}{space 1}      160{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       59{col 38}{space 1}       10{col 50}{result}{space 1}       49{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est5{txt} stored)

{com}. 
. reghdfe log_lev ss_pre_all_debt_banintsty_n1_sd if coal_industry_mining==1, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 155 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 13 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     1,189
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    155{txt}){col 67}= {res}      3.90
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0500
{txt}{col 51}R-squared{col 67}= {res}    0.7406
{txt}{col 51}Adj R-squared{col 67}= {res}    0.6157
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0073
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       156{txt}{col 51}Root MSE{col 67}= {res}    0.7565

{txt}{ralign 97:(Std. err. adjusted for {res:156} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                        log_lev{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.1856637{col 45}{space 2} .0939749{col 56}{space 1}   -1.98{col 65}{space 3}0.050{col 73}{space 4}-.3713005{col 86}{space 3}-.0000269
{txt}{space 26}_cons {c |}{col 33}{res}{space 2}-1.582272{col 45}{space 2} .0289918{col 56}{space 1}  -54.58{col 65}{space 3}0.000{col 73}{space 4}-1.639542{col 86}{space 3}-1.525002
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      156{col 38}{space 1}      156{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      181{col 38}{space 1}       10{col 50}{result}{space 1}      171{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       59{col 38}{space 1}       10{col 50}{result}{space 1}       49{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est6{txt} stored)

{com}. 
. esttab, label starlevels(* .10 ** .05 *** .01) nocons
{res}
{txt}{hline 116}
{txt}                              (1)             (2)             (3)             (4)             (5)             (6)   
{txt}                          log_lev         log_lev         log_lev         log_lev         log_lev         log_lev   
{txt}{hline 116}
{txt}$\text{Bank Exit E~}{res}      -0.0941**        -0.247***                                       -0.274**        -0.186** {txt}
                    {res} {ralign 12:{txt:(}-2.16{txt:)}}    {ralign 12:{txt:(}-3.03{txt:)}}                                    {ralign 12:{txt:(}-2.03{txt:)}}    {ralign 12:{txt:(}-1.98{txt:)}}   {txt}

{txt}Low Coal Share # $~i{res}                                       -0.287***                                                {txt}
                    {res}                                 {ralign 12:{txt:(}-3.05{txt:)}}                                                   {txt}

{txt}High Coal Share # ~x{res}                                       -0.221**                                                 {txt}
                    {res}                                 {ralign 12:{txt:(}-2.14{txt:)}}                                                   {txt}

{txt}Small Firm # $\tex~x{res}                                                       -0.285***                                {txt}
                    {res}                                                 {ralign 12:{txt:(}-2.82{txt:)}}                                   {txt}

{txt}Large Firm # $\tex~x{res}                                                       -0.142***                                {txt}
                    {res}                                                 {ralign 12:{txt:(}-2.98{txt:)}}                                   {txt}
{txt}{hline 116}
{txt}Observations        {res}         2530            2366            2173            2366            1070            1189   {txt}
{txt}{hline 116}
{txt}t statistics in parentheses
{txt}* p<.10, ** p<.05, *** p<.01

{com}. 
. esttab using "../Results/Tables/Table_8_Panel_B.tex", replace booktabs b(%8.3f) se(%8.3f)  ///
>                 nocons nonotes starlevels(* .10 ** .05 *** .01) nolegend numbers ///
>                 label interaction(" $\times$ ") ///
>                                 substitute("\_" "_") ///
>                 nomtitles prehead("{c -(} \def\sym#1{c -(}\ifmmode^{c -(}#1{c )-}\else\(^{c -(}#1{c )-}\)\fi{c )-}" ///
>                                                               "\begin{c -(}tabular{c )-}{c -(}l*{c -(}@M{c )-}{c -(}r{c )-}{c )-}" ///
>                                                               "\toprule" ///
>                                                               "&  \multicolumn{c -(}6{c )-}{c -(}c{c )-}{c -(}\textbf{c -(}Leverage (log){c )-}{c )-}  \\" ///
>                                                                                                                           "\cmidrule{c -(}2-7{c )-} & & & & & Power & Mining \\" ///
>                                                               "\cmidrule(l{c -(}2pt{c )-}){c -(}6-7{c )-}") ///
>                 stats(borrower y cy sy N r2_a , fmt(0 0 0 0 %8.0fc 3) labels("Borrower FE" "Year FE" "Country x Year FE" "Size x Year FE" "Observations" "Adj-R$^2$" ))
{res}{txt}(output written to {browse  `"../Results/Tables/Table_8_Panel_B.tex"'})

{com}.         
. ***Panel C: Total Assets  
. eststo clear
{txt}
{com}. 
. reghdfe log_size ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year) cl(borrower_id)
{res}{txt}(dropped 6 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,786
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}    346{txt}){col 67}= {res}      6.44
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0116
{txt}{col 51}R-squared{col 67}= {res}    0.9754
{txt}{col 51}Adj R-squared{col 67}= {res}    0.9717
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0083
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       347{txt}{col 51}Root MSE{col 67}= {res}    0.3673

{txt}{ralign 97:(Std. err. adjusted for {res:347} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                       log_size{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.0568615{col 45}{space 2} .0224121{col 56}{space 1}   -2.54{col 65}{space 3}0.012{col 73}{space 4}-.1009426{col 86}{space 3}-.0127805
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 8.431037{col 45}{space 2} .0085879{col 56}{space 1}  981.74{col 65}{space 3}0.000{col 73}{space 4} 8.414146{col 86}{space 3} 8.447928
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 13}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text} Absorbed FE{col 14}{c |} Categories{col 27} - Redundant{col 39}  = Num. Coefs{col 54}{c |}
{res}{col 1}{text}{hline 13}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text} borrower_id{col 14}{c |}{space 1}      347{col 27}{space 1}      347{col 39}{result}{space 1}        0{col 53}{text}*{col 54}{c |}
{res}{col 1}{text}        year{col 14}{c |}{space 1}       10{col 27}{space 1}        0{col 39}{result}{space 1}       10{col 53}{text} {col 54}{c |}
{res}{col 1}{text}{hline 13}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE4

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:No}"

added macro:
                 e(sy) : "{res:No}"

{com}. eststo
{txt}({res}est1{txt} stored)

{com}. 
. reghdfe log_size ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 195 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 12 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,597
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    323{txt}){col 67}= {res}     10.00
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0017
{txt}{col 51}R-squared{col 67}= {res}    0.9803
{txt}{col 51}Adj R-squared{col 67}= {res}    0.9739
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0128
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       324{txt}{col 51}Root MSE{col 67}= {res}    0.3549

{txt}{ralign 97:(Std. err. adjusted for {res:324} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                       log_size{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.0913284{col 45}{space 2} .0288749{col 56}{space 1}   -3.16{col 65}{space 3}0.002{col 73}{space 4} -.148135{col 86}{space 3}-.0345217
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 8.489567{col 45}{space 2}   .01055{col 56}{space 1}  804.70{col 65}{space 3}0.000{col 73}{space 4} 8.468812{col 86}{space 3} 8.510322
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      324{col 38}{space 1}      324{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      265{col 38}{space 1}       10{col 50}{result}{space 1}      255{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       60{col 38}{space 1}       10{col 50}{result}{space 1}       50{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est2{txt} stored)

{com}. 
. reghdfe log_size i.highshare#c.ss_pre_all_debt_banintsty_n1_sd, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 175 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 11 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,366
{txt}Absorbing 4 HDFE groups{col 51}F({res}   2{txt},{res}    294{txt}){col 67}= {res}      6.21
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0023
{txt}{col 51}R-squared{col 67}= {res}    0.9841
{txt}{col 51}Adj R-squared{col 67}= {res}    0.9785
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0233
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       295{txt}{col 51}Root MSE{col 67}= {res}    0.3110

{txt}{ralign 109:(Std. err. adjusted for {res:295} clusters in {res:borrower_id})}
{hline 44}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 45}{c |}{col 57}    Robust
{col 1}                                   log_size{col 45}{c |} Coefficient{col 57}  std. err.{col 69}      t{col 77}   P>|t|{col 85}     [95% con{col 98}f. interval]
{hline 44}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
highshare#c.ss_pre_all_debt_banintsty_n1_sd {c |}
{space 28}Low Coal Share  {c |}{col 45}{res}{space 2} -.086275{col 57}{space 2} .0283921{col 68}{space 1}   -3.04{col 77}{space 3}0.003{col 85}{space 4}-.1421524{col 98}{space 3}-.0303975
{txt}{space 27}High Coal Share  {c |}{col 45}{res}{space 2}-.1431901{col 57}{space 2} .0449174{col 68}{space 1}   -3.19{col 77}{space 3}0.002{col 85}{space 4}-.2315906{col 98}{space 3}-.0547897
{txt}{space 43} {c |}
{space 38}_cons {c |}{col 45}{res}{space 2} 8.568875{col 57}{space 2} .0115861{col 68}{space 1}  739.58{col 77}{space 3}0.000{col 85}{space 4} 8.546072{col 98}{space 3} 8.591677
{txt}{hline 44}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      295{col 38}{space 1}      295{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      265{col 38}{space 1}       10{col 50}{result}{space 1}      255{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       60{col 38}{space 1}       10{col 50}{result}{space 1}       50{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est3{txt} stored)

{com}. 
. reghdfe log_size c.ss_pre_all_debt_banintsty_n1_sd#i.median_assets_mean, ///
>     absorb(borrower_id i.count#i.year i.year) cl(borrower_id)
{res}{txt}(dropped 195 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 9 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     2,597
{txt}Absorbing 3 HDFE groups{col 51}F({res}   2{txt},{res}    323{txt}){col 67}= {res}      6.52
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0017
{txt}{col 51}R-squared{col 67}= {res}    0.9796
{txt}{col 51}Adj R-squared{col 67}= {res}    0.9736
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0201
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       324{txt}{col 51}Root MSE{col 67}= {res}    0.3570

{txt}{ralign 118:(Std. err. adjusted for {res:324} clusters in {res:borrower_id})}
{hline 53}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 54}{c |}{col 66}    Robust
{col 1}                                            log_size{col 54}{c |} Coefficient{col 66}  std. err.{col 78}      t{col 86}   P>|t|{col 94}     [95% con{col 107}f. interval]
{hline 53}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
median_assets_mean#c.ss_pre_all_debt_banintsty_n1_sd {c |}
{space 41}Small Firm  {c |}{col 54}{res}{space 2}-.1152204{col 66}{space 2} .0416631{col 77}{space 1}   -2.77{col 86}{space 3}0.006{col 94}{space 4}-.1971856{col 107}{space 3}-.0332552
{txt}{space 41}Large Firm  {c |}{col 54}{res}{space 2}-.0881372{col 66}{space 2} .0272104{col 77}{space 1}   -3.24{col 86}{space 3}0.001{col 94}{space 4}-.1416693{col 107}{space 3}-.0346052
{txt}{space 52} {c |}
{space 47}_cons {c |}{col 54}{res}{space 2} 8.491838{col 66}{space 2} .0098863{col 77}{space 1}  858.95{col 86}{space 3}0.000{col 94}{space 4} 8.472388{col 107}{space 3} 8.511287
{txt}{hline 53}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 14}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}  Absorbed FE{col 15}{c |} Categories{col 28} - Redundant{col 40}  = Num. Coefs{col 55}{c |}
{res}{col 1}{text}{hline 14}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}  borrower_id{col 15}{c |}{space 1}      324{col 28}{space 1}      324{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}   count#year{col 15}{c |}{space 1}      265{col 28}{space 1}        0{col 40}{result}{space 1}      265{col 54}{text} {col 55}{c |}
{res}{col 1}{text}         year{col 15}{c |}{space 1}       10{col 28}{space 1}       10{col 40}{result}{space 1}        0{col 54}{text} {col 55}{c |}
{res}{col 1}{text}{hline 14}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3nosy

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:No}"

{com}. eststo
{txt}({res}est4{txt} stored)

{com}. 
. reghdfe log_size ss_pre_all_debt_banintsty_n1_sd if coal_industry_mining==0, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 144 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 12 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     1,158
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    147{txt}){col 67}= {res}      3.65
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0580
{txt}{col 51}R-squared{col 67}= {res}    0.9792
{txt}{col 51}Adj R-squared{col 67}= {res}    0.9691
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0116
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       148{txt}{col 51}Root MSE{col 67}= {res}    0.3595

{txt}{ralign 97:(Std. err. adjusted for {res:148} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                       log_size{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.0910468{col 45}{space 2} .0476614{col 56}{space 1}   -1.91{col 65}{space 3}0.058{col 73}{space 4}-.1852368{col 86}{space 3} .0031432
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 8.741882{col 45}{space 2} .0197655{col 56}{space 1}  442.28{col 65}{space 3}0.000{col 73}{space 4} 8.702821{col 86}{space 3} 8.780944
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      148{col 38}{space 1}      148{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      180{col 38}{space 1}       10{col 50}{result}{space 1}      170{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       59{col 38}{space 1}       10{col 50}{result}{space 1}       49{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est5{txt} stored)

{com}. 
. reghdfe log_size ss_pre_all_debt_banintsty_n1_sd if coal_industry_mining==1, ///
>     absorb(borrower_id year i.count#i.year quintile_assets#i.year) ///
>     cl(borrower_id)
{res}{txt}(dropped 150 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 12 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     1,340
{txt}Absorbing 4 HDFE groups{col 51}F({res}   1{txt},{res}    165{txt}){col 67}= {res}      6.13
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0143
{txt}{col 51}R-squared{col 67}= {res}    0.9834
{txt}{col 51}Adj R-squared{col 67}= {res}    0.9759
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0136
{txt}{col 1}Number of clusters ({res}borrower_id{txt}) {col 30}= {res}       166{txt}{col 51}Root MSE{col 67}= {res}    0.3552

{txt}{ralign 97:(Std. err. adjusted for {res:166} clusters in {res:borrower_id})}
{hline 32}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 33}{c |}{col 45}    Robust
{col 1}                       log_size{col 33}{c |} Coefficient{col 45}  std. err.{col 57}      t{col 65}   P>|t|{col 73}     [95% con{col 86}f. interval]
{hline 32}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ss_pre_all_debt_banintsty_n1_sd {c |}{col 33}{res}{space 2}-.0951374{col 45}{space 2} .0384121{col 56}{space 1}   -2.48{col 65}{space 3}0.014{col 73}{space 4}-.1709801{col 86}{space 3}-.0192947
{txt}{space 26}_cons {c |}{col 33}{res}{space 2} 8.272281{col 45}{space 2} .0122917{col 56}{space 1}  673.00{col 65}{space 3}0.000{col 73}{space 4} 8.248012{col 86}{space 3}  8.29655
{txt}{hline 32}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 24}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}            Absorbed FE{col 25}{c |} Categories{col 38} - Redundant{col 50}  = Num. Coefs{col 65}{c |}
{res}{col 1}{text}{hline 24}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}            borrower_id{col 25}{c |}{space 1}      166{col 38}{space 1}      166{col 50}{result}{space 1}        0{col 64}{text}*{col 65}{c |}
{res}{col 1}{text}                   year{col 25}{c |}{space 1}       10{col 38}{space 1}        0{col 50}{result}{space 1}       10{col 64}{text} {col 65}{c |}
{res}{col 1}{text}             count#year{col 25}{c |}{space 1}      198{col 38}{space 1}       10{col 50}{result}{space 1}      188{col 64}{text} {col 65}{c |}
{res}{col 1}{text}   quintile_assets#year{col 25}{c |}{space 1}       59{col 38}{space 1}       10{col 50}{result}{space 1}       49{col 64}{text}?{col 65}{c |}
{res}{col 1}{text}{hline 24}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
? = number of redundant parameters may be higher
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE3

{txt}added macro:
           e(borrower) : "{res:Yes}"

added macro:
                  e(y) : "{res:Yes}"

added macro:
                 e(cy) : "{res:Yes}"

added macro:
                 e(sy) : "{res:Yes}"

{com}. eststo
{txt}({res}est6{txt} stored)

{com}. 
. esttab, label starlevels(* .10 ** .05 *** .01) nocons
{res}
{txt}{hline 116}
{txt}                              (1)             (2)             (3)             (4)             (5)             (6)   
{txt}                         log_size        log_size        log_size        log_size        log_size        log_size   
{txt}{hline 116}
{txt}$\text{Bank Exit E~}{res}      -0.0569**       -0.0913***                                      -0.0910*        -0.0951** {txt}
                    {res} {ralign 12:{txt:(}-2.54{txt:)}}    {ralign 12:{txt:(}-3.16{txt:)}}                                    {ralign 12:{txt:(}-1.91{txt:)}}    {ralign 12:{txt:(}-2.48{txt:)}}   {txt}

{txt}Low Coal Share # $~i{res}                                      -0.0863***                                                {txt}
                    {res}                                 {ralign 12:{txt:(}-3.04{txt:)}}                                                   {txt}

{txt}High Coal Share # ~x{res}                                       -0.143***                                                {txt}
                    {res}                                 {ralign 12:{txt:(}-3.19{txt:)}}                                                   {txt}

{txt}Small Firm # $\tex~x{res}                                                       -0.115***                                {txt}
                    {res}                                                 {ralign 12:{txt:(}-2.77{txt:)}}                                   {txt}

{txt}Large Firm # $\tex~x{res}                                                      -0.0881***                                {txt}
                    {res}                                                 {ralign 12:{txt:(}-3.24{txt:)}}                                   {txt}
{txt}{hline 116}
{txt}Observations        {res}         2786            2597            2366            2597            1158            1340   {txt}
{txt}{hline 116}
{txt}t statistics in parentheses
{txt}* p<.10, ** p<.05, *** p<.01

{com}. 
. esttab using "../Results/Tables/Table_8_Panel_C.tex", replace booktabs b(%8.3f) se(%8.3f)  ///
>                 nocons nonotes starlevels(* .10 ** .05 *** .01) nolegend numbers ///
>                 label interaction(" $\times$ ") ///
>                                 substitute("\_" "_") ///
>                 nomtitles prehead("{c -(} \def\sym#1{c -(}\ifmmode^{c -(}#1{c )-}\else\(^{c -(}#1{c )-}\)\fi{c )-}" ///
>                                                               "\begin{c -(}tabular{c )-}{c -(}l*{c -(}@M{c )-}{c -(}r{c )-}{c )-}" ///
>                                                               "\toprule" ///
>                                                               "&  \multicolumn{c -(}6{c )-}{c -(}c{c )-}{c -(}\textbf{c -(}Total Assets (log){c )-}{c )-}  \\" ///
>                                                                                                                           "\cmidrule{c -(}2-7{c )-} & & & & & Power & Mining \\" ///
>                                                               "\cmidrule(l{c -(}2pt{c )-}){c -(}6-7{c )-}") ///
>                 stats(borrower y cy sy N r2_a , fmt(0 0 0 0 %8.0fc 3) labels("Borrower FE" "Year FE" "Country x Year FE" "Size x Year FE" "Observations" "Adj-R$^2$" ))
{res}{txt}(output written to {browse  `"../Results/Tables/Table_8_Panel_C.tex"'})

{com}.                                 
. ********************************************************************************
. ********************************************************************************
. * Table 9: Facility Level Effects
. ********************************************************************************
. ********************************************************************************
. //Panel A: Effects on Coal-fired Power Plant Closures
. use ../Intermediate/plant_year_panel_clean, clear
{txt}
{com}. 
. *Setting up Cox Model
. stset year, failure(retired) id(plant_parent_id) origin(first_date_Operating) 

{txt}Survival-time data settings

{col 12}ID variable: {res}plant_parent_id
{col 10}{txt}Failure event: {res}retired!=0 & retired<.
{col 1}{txt}Observed time interval: {res}(year[_n-1], year]
{col 6}{txt}Exit on or before: {res}failure
{col 6}{txt}Time for analysis: {res}(time-origin)
{col 17}{txt}Origin: {res}time first_date_Operating_year

{txt}{hline 74}
{res}     55,770{txt}  total observations
{res}     10,725{txt}  ignored because never entered
{res}      7,225{txt}  observations end on or before {bf:enter()}
{res}      7,249{txt}  observations begin on or after (first) failure
{hline 74}
{res}     30,571{txt}  observations remaining, representing
{res}      3,285{txt}  subjects
{res}      1,071{txt}  failures in single-failure-per-subject data
{res}    102,670{txt}  total analysis time at risk and under observation
                                                At risk from t = {res}        0
                                     {txt}Earliest observed entry t = {res}        0
                                          {txt}Last observed exit t = {res}       74
{txt}
{com}. local clvar Country
{txt}
{com}. 
. eststo clear
{txt}
{com}. 
. stcox ban_intensity_max_sd if year < 2015, strata(Country) vce(cl ParentID_GCEL)

{col 9}{txt}Failure {bf:_d}: {res}retired
{col 3}{txt}Analysis time {bf:_t}: {res}(year-origin)
{col 13}{txt}Origin: {res}time first_date_Operating_year
{col 8}{txt}ID variable: {res}plant_parent_id

{txt}Iteration 0:  Log pseudolikelihood = {res}-2220.8123
{txt}Iteration 1:  Log pseudolikelihood = {res}-2220.8118
{txt}Iteration 2:  Log pseudolikelihood = {res}-2220.8118
{txt}Refining estimates:
Iteration 0:  Log pseudolikelihood = {res}-2220.8118

{txt}Stratified Cox regression with Breslow method for ties
Strata variable: {res:Country}

No. of subjects = {res}{ralign 6:2,853}{col 57}{txt}{lalign 13:Number of obs} = {res}{ralign 6:14,225}
{txt}No. of failures = {res}{ralign 6:503}
{txt}Time at risk    = {res}{ralign 6:86,324}
{col 57}{txt}{lalign 13:Wald chi2({res:1})} = {res}{ralign 6:0.00}
{txt}Log pseudolikelihood = {res}-2220.8118{col 57}{txt}{lalign 13:Prob > chi2} = {res}{ralign 6:0.9858}

{txt}{ralign 86:(Std. err. adjusted for {res:206} clusters in {res:ParentID_GCEL})}
{hline 21}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 22}{c |}{col 34}    Robust
{col 1}                  _t{col 22}{c |} Haz. ratio{col 34}   std. err.{col 46}      z{col 54}   P>|z|{col 62}     [95% con{col 75}f. interval]
{hline 21}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
ban_intensity_max_sd {c |}{col 22}{res}{space 2} .9953244{col 34}{space 2} .2626548{col 45}{space 1}   -0.02{col 54}{space 3}0.986{col 62}{space 4} .5933927{col 75}{space 3} 1.669502
{txt}{hline 21}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. FE_cy 

{txt}added macro:
            e(country) : "{res:Yes}"

{com}. eststo 
{txt}({res}est1{txt} stored)

{com}. 
. stcox c.post2015##c.ban_intensity_max_sd, strata(country_i) vce(cl ParentID_GCEL) nohr

{col 9}{txt}Failure {bf:_d}: {res}retired
{col 3}{txt}Analysis time {bf:_t}: {res}(year-origin)
{col 13}{txt}Origin: {res}time first_date_Operating_year
{col 8}{txt}ID variable: {res}plant_parent_id

{txt}Iteration 0:  Log pseudolikelihood = {res}-4995.7506
{txt}Iteration 1:  Log pseudolikelihood = {res}-4840.6747
{txt}Iteration 2:  Log pseudolikelihood = {res}-4838.8087
{txt}Iteration 3:  Log pseudolikelihood = {res}-4838.8087
{txt}Refining estimates:
Iteration 0:  Log pseudolikelihood = {res}-4838.8087

{txt}Stratified Cox regression with Breslow method for ties
Strata variable: {res:country_i}

No. of subjects = {res}{ralign 7:3,285}{col 57}{txt}{lalign 13:Number of obs} = {res}{ralign 6:30,571}
{txt}No. of failures = {res}{ralign 7:1,071}
{txt}Time at risk    = {res}{ralign 7:102,670}
{col 57}{txt}{lalign 13:Wald chi2({res:3})} = {res}{ralign 6:51.44}
{txt}Log pseudolikelihood = {res}-4838.8087{col 57}{txt}{lalign 13:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 99:(Std. err. adjusted for {res:237} clusters in {res:ParentID_GCEL})}
{hline 34}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 35}{c |}{col 47}    Robust
{col 1}                               _t{col 35}{c |} Coefficient{col 47}  std. err.{col 59}      z{col 67}   P>|z|{col 75}     [95% con{col 88}f. interval]
{hline 34}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 25}post2015 {c |}{col 35}{res}{space 2} .6989231{col 47}{space 2}  .195415{col 58}{space 1}    3.58{col 67}{space 3}0.000{col 75}{space 4} .3159167{col 88}{space 3} 1.081929
{txt}{space 13}ban_intensity_max_sd {c |}{col 35}{res}{space 2}-.1957898{col 47}{space 2} .1358882{col 58}{space 1}   -1.44{col 67}{space 3}0.150{col 75}{space 4}-.4621257{col 88}{space 3} .0705462
{txt}{space 33} {c |}
c.post2015#c.ban_intensity_max_sd {c |}{col 35}{res}{space 2} .3428632{col 47}{space 2} .1402597{col 58}{space 1}    2.44{col 67}{space 3}0.015{col 75}{space 4} .0679592{col 88}{space 3} .6177672
{txt}{hline 34}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. FE_cy 

{txt}added macro:
            e(country) : "{res:Yes}"

{com}. eststo 
{txt}({res}est2{txt} stored)

{com}. 
. stcox c.post2015##c.ban_intensity_max_sd##c.small_firm, strata(Country) vce(cl ParentID_GCEL) 

{col 9}{txt}Failure {bf:_d}: {res}retired
{col 3}{txt}Analysis time {bf:_t}: {res}(year-origin)
{col 13}{txt}Origin: {res}time first_date_Operating_year
{col 8}{txt}ID variable: {res}plant_parent_id

{txt}Iteration 0:  Log pseudolikelihood = {res}-4995.7506
{txt}Iteration 1:  Log pseudolikelihood = {res}-4828.7468
{txt}Iteration 2:  Log pseudolikelihood = {res} -4820.995
{txt}Iteration 3:  Log pseudolikelihood = {res}-4819.7655
{txt}Iteration 4:  Log pseudolikelihood = {res} -4819.495
{txt}Iteration 5:  Log pseudolikelihood = {res}-4819.4708
{txt}Iteration 6:  Log pseudolikelihood = {res}-4819.4705
{txt}Iteration 7:  Log pseudolikelihood = {res}-4819.4705
{txt}Refining estimates:
Iteration 0:  Log pseudolikelihood = {res}-4819.4705

{txt}Stratified Cox regression with Breslow method for ties
Strata variable: {res:Country}

No. of subjects = {res}{ralign 7:3,285}{col 57}{txt}{lalign 13:Number of obs} = {res}{ralign 6:30,571}
{txt}No. of failures = {res}{ralign 7:1,071}
{txt}Time at risk    = {res}{ralign 7:102,670}
{col 57}{txt}{lalign 13:Wald chi2({res:7})} = {res}{ralign 6:79.31}
{txt}Log pseudolikelihood = {res}-4819.4705{col 57}{txt}{lalign 13:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 112:(Std. err. adjusted for {res:237} clusters in {res:ParentID_GCEL})}
{hline 47}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 48}{c |}{col 60}    Robust
{col 1}                                            _t{col 48}{c |} Haz. ratio{col 60}   std. err.{col 72}      z{col 80}   P>|z|{col 88}     [95% con{col 101}f. interval]
{hline 47}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 38}post2015 {c |}{col 48}{res}{space 2} 1.643257{col 60}{space 2} .3217508{col 71}{space 1}    2.54{col 80}{space 3}0.011{col 88}{space 4}  1.11954{col 101}{space 3} 2.411966
{txt}{space 26}ban_intensity_max_sd {c |}{col 48}{res}{space 2} .8133071{col 60}{space 2} .1232834{col 71}{space 1}   -1.36{col 80}{space 3}0.173{col 88}{space 4} .6042644{col 101}{space 3} 1.094667
{txt}{space 46} {c |}
{space 13}c.post2015#c.ban_intensity_max_sd {c |}{col 48}{res}{space 2} 1.493363{col 60}{space 2} .2144821{col 71}{space 1}    2.79{col 80}{space 3}0.005{col 88}{space 4} 1.126971{col 101}{space 3} 1.978873
{txt}{space 46} {c |}
{space 36}small_firm {c |}{col 48}{res}{space 2} .6283503{col 60}{space 2} .3745219{col 71}{space 1}   -0.78{col 80}{space 3}0.436{col 88}{space 4} .1953673{col 101}{space 3} 2.020933
{txt}{space 46} {c |}
{space 23}c.post2015#c.small_firm {c |}{col 48}{res}{space 2} 2.947894{col 60}{space 2}  1.63731{col 71}{space 1}    1.95{col 80}{space 3}0.052{col 88}{space 4} .9925227{col 101}{space 3}  8.75555
{txt}{space 46} {c |}
{space 11}c.ban_intensity_max_sd#c.small_firm {c |}{col 48}{res}{space 2} .2244304{col 60}{space 2} .1690596{col 71}{space 1}   -1.98{col 80}{space 3}0.047{col 88}{space 4} .0512727{col 101}{space 3} .9823751
{txt}{space 46} {c |}
c.post2015#c.ban_intensity_max_sd#c.small_firm {c |}{col 48}{res}{space 2}  4.12258{col 60}{space 2} 3.031952{col 71}{space 1}    1.93{col 80}{space 3}0.054{col 88}{space 4} .9753329{col 101}{space 3} 17.42551
{txt}{hline 47}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. FE_cy 

{txt}added macro:
            e(country) : "{res:Yes}"

{com}. eststo 
{txt}({res}est3{txt} stored)

{com}. 
. stcox c.post2015##c.ban_intensity_max_sd##c.share_low, strata(Country)  vce(cl ParentID_GCEL) 

{col 9}{txt}Failure {bf:_d}: {res}retired
{col 3}{txt}Analysis time {bf:_t}: {res}(year-origin)
{col 13}{txt}Origin: {res}time first_date_Operating_year
{col 8}{txt}ID variable: {res}plant_parent_id

{txt}Iteration 0:  Log pseudolikelihood = {res}-4912.7008
{txt}Iteration 1:  Log pseudolikelihood = {res}-4756.1109
{txt}Iteration 2:  Log pseudolikelihood = {res}-4752.7832
{txt}Iteration 3:  Log pseudolikelihood = {res}-4752.7795
{txt}Refining estimates:
Iteration 0:  Log pseudolikelihood = {res}-4752.7795

{txt}Stratified Cox regression with Breslow method for ties
Strata variable: {res:Country}

No. of subjects = {res}{ralign 7:3,210}{col 57}{txt}{lalign 13:Number of obs} = {res}{ralign 6:29,777}
{txt}No. of failures = {res}{ralign 7:1,052}
{txt}Time at risk    = {res}{ralign 7:100,821}
{col 57}{txt}{lalign 13:Wald chi2({res:7})} = {res}{ralign 6:65.56}
{txt}Log pseudolikelihood = {res}-4752.7795{col 57}{txt}{lalign 13:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 111:(Std. err. adjusted for {res:229} clusters in {res:ParentID_GCEL})}
{hline 46}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 47}{c |}{col 59}    Robust
{col 1}                                           _t{col 47}{c |} Haz. ratio{col 59}   std. err.{col 71}      z{col 79}   P>|z|{col 87}     [95% con{col 100}f. interval]
{hline 46}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 37}post2015 {c |}{col 47}{res}{space 2} 2.340622{col 59}{space 2} .4878085{col 70}{space 1}    4.08{col 79}{space 3}0.000{col 87}{space 4} 1.555724{col 100}{space 3} 3.521519
{txt}{space 25}ban_intensity_max_sd {c |}{col 47}{res}{space 2} .7944043{col 59}{space 2} .1879693{col 70}{space 1}   -0.97{col 79}{space 3}0.331{col 87}{space 4} .4996123{col 100}{space 3} 1.263136
{txt}{space 45} {c |}
{space 12}c.post2015#c.ban_intensity_max_sd {c |}{col 47}{res}{space 2} 1.329261{col 59}{space 2} .3040548{col 70}{space 1}    1.24{col 79}{space 3}0.213{col 87}{space 4} .8489985{col 100}{space 3} 2.081201
{txt}{space 45} {c |}
{space 36}share_low {c |}{col 47}{res}{space 2} .8887082{col 59}{space 2} .1497013{col 70}{space 1}   -0.70{col 79}{space 3}0.484{col 87}{space 4} .6388161{col 100}{space 3} 1.236353
{txt}{space 45} {c |}
{space 23}c.post2015#c.share_low {c |}{col 47}{res}{space 2} .4642194{col 59}{space 2} .1852036{col 70}{space 1}   -1.92{col 79}{space 3}0.054{col 87}{space 4} .2123882{col 100}{space 3}  1.01465
{txt}{space 45} {c |}
{space 11}c.ban_intensity_max_sd#c.share_low {c |}{col 47}{res}{space 2} 1.024759{col 59}{space 2} .2152954{col 70}{space 1}    0.12{col 79}{space 3}0.907{col 87}{space 4} .6788751{col 100}{space 3}  1.54687
{txt}{space 45} {c |}
c.post2015#c.ban_intensity_max_sd#c.share_low {c |}{col 47}{res}{space 2} 1.352792{col 59}{space 2} .3653011{col 70}{space 1}    1.12{col 79}{space 3}0.263{col 87}{space 4} .7968508{col 100}{space 3} 2.296599
{txt}{hline 46}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. FE_cy 

{txt}added macro:
            e(country) : "{res:Yes}"

{com}. eststo 
{txt}({res}est4{txt} stored)

{com}. 
. stcox c.post2015##c.ban_intensity_max_sd##c.big_plant, strata(Country)  vce(cl ParentID_GCEL) 

{col 9}{txt}Failure {bf:_d}: {res}retired
{col 3}{txt}Analysis time {bf:_t}: {res}(year-origin)
{col 13}{txt}Origin: {res}time first_date_Operating_year
{col 8}{txt}ID variable: {res}plant_parent_id

{txt}Iteration 0:  Log pseudolikelihood = {res}-4995.7506
{txt}Iteration 1:  Log pseudolikelihood = {res}-4823.3954
{txt}Iteration 2:  Log pseudolikelihood = {res}-4819.2318
{txt}Iteration 3:  Log pseudolikelihood = {res}-4819.1353
{txt}Iteration 4:  Log pseudolikelihood = {res}-4819.1335
{txt}Iteration 5:  Log pseudolikelihood = {res}-4819.1335
{txt}Refining estimates:
Iteration 0:  Log pseudolikelihood = {res}-4819.1335

{txt}Stratified Cox regression with Breslow method for ties
Strata variable: {res:Country}

No. of subjects = {res}{ralign 7:3,285}{col 57}{txt}{lalign 13:Number of obs} = {res}{ralign 6:30,571}
{txt}No. of failures = {res}{ralign 7:1,071}
{txt}Time at risk    = {res}{ralign 7:102,670}
{col 57}{txt}{lalign 13:Wald chi2({res:7})} = {res}{ralign 6:65.23}
{txt}Log pseudolikelihood = {res}-4819.1335{col 57}{txt}{lalign 13:Prob > chi2} = {res}{ralign 6:0.0000}

{txt}{ralign 111:(Std. err. adjusted for {res:237} clusters in {res:ParentID_GCEL})}
{hline 46}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 47}{c |}{col 59}    Robust
{col 1}                                           _t{col 47}{c |} Haz. ratio{col 59}   std. err.{col 71}      z{col 79}   P>|z|{col 87}     [95% con{col 100}f. interval]
{hline 46}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 37}post2015 {c |}{col 47}{res}{space 2} 2.228458{col 59}{space 2} .4412948{col 70}{space 1}    4.05{col 79}{space 3}0.000{col 87}{space 4} 1.511623{col 100}{space 3} 3.285227
{txt}{space 25}ban_intensity_max_sd {c |}{col 47}{res}{space 2}  .826371{col 59}{space 2} .1152869{col 70}{space 1}   -1.37{col 79}{space 3}0.172{col 87}{space 4} .6286719{col 100}{space 3} 1.086241
{txt}{space 45} {c |}
{space 12}c.post2015#c.ban_intensity_max_sd {c |}{col 47}{res}{space 2} 1.365174{col 59}{space 2} .2037445{col 70}{space 1}    2.09{col 79}{space 3}0.037{col 87}{space 4} 1.018946{col 100}{space 3} 1.829048
{txt}{space 45} {c |}
{space 36}big_plant {c |}{col 47}{res}{space 2} .1179236{col 59}{space 2} .1074395{col 70}{space 1}   -2.35{col 79}{space 3}0.019{col 87}{space 4} .0197732{col 100}{space 3}  .703275
{txt}{space 45} {c |}
{space 23}c.post2015#c.big_plant {c |}{col 47}{res}{space 2} 1.642371{col 59}{space 2} 1.599542{col 70}{space 1}    0.51{col 79}{space 3}0.610{col 87}{space 4} .2434819{col 100}{space 3} 11.07837
{txt}{space 45} {c |}
{space 11}c.ban_intensity_max_sd#c.big_plant {c |}{col 47}{res}{space 2} 1.524775{col 59}{space 2} .3069531{col 70}{space 1}    2.10{col 79}{space 3}0.036{col 87}{space 4} 1.027661{col 100}{space 3} 2.262359
{txt}{space 45} {c |}
c.post2015#c.ban_intensity_max_sd#c.big_plant {c |}{col 47}{res}{space 2} 1.167681{col 59}{space 2} .2927571{col 70}{space 1}    0.62{col 79}{space 3}0.536{col 87}{space 4} .7143549{col 100}{space 3} 1.908687
{txt}{hline 46}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}

{com}. FE_cy 

{txt}added macro:
            e(country) : "{res:Yes}"

{com}. eststo 
{txt}({res}est5{txt} stored)

{com}. 
. esttab, starlevels(* .10 ** .05 *** .01) eform t(%8.3f) label drop(c.ban_intensity_max_sd#c.small_firm c.ban_intensity_max_sd#c.share_low c.ban_intensity_max_sd#c.big_plant post2015 big_plant small_firm c.post2015#c.big_plant c.post2015#c.small_firm  c.post2015#c.share_low share_low) interaction(" $\times$ ")
{res}
{txt}{hline 100}
{txt}                              (1)             (2)             (3)             (4)             (5)   
{txt}                     Analysis t~s    Analysis t~s    Analysis t~s    Analysis t~s    Analysis t~s   
{txt}{hline 100}
{txt}$\text{Bank Exit E~}{res}        0.995           0.822           0.813           0.794           0.826   {txt}
                    {res} {ralign 12:{txt:(}-0.018{txt:)}}    {ralign 12:{txt:(}-1.441{txt:)}}    {ralign 12:{txt:(}-1.363{txt:)}}    {ralign 12:{txt:(}-0.973{txt:)}}    {ralign 12:{txt:(}-1.367{txt:)}}   {txt}

{txt}$ Year \geq 2015$ ~x{res}                        1.409**         1.493***        1.329           1.365** {txt}
                    {res}                 {ralign 12:{txt:(}2.444{txt:)}}    {ralign 12:{txt:(}2.792{txt:)}}    {ralign 12:{txt:(}1.244{txt:)}}    {ralign 12:{txt:(}2.086{txt:)}}   {txt}

{txt}$ Year \geq 2015$ ~x{res}                                        4.123*                                  {txt}
                    {res}                                 {ralign 12:{txt:(}1.926{txt:)}}                                   {txt}

{txt}$ Year \geq 2015$ ~x{res}                                                        1.353                   {txt}
                    {res}                                                 {ralign 12:{txt:(}1.119{txt:)}}                   {txt}

{txt}$ Year \geq 2015$ ~x{res}                                                                        1.168   {txt}
                    {res}                                                                 {ralign 12:{txt:(}0.618{txt:)}}   {txt}
{txt}{hline 100}
{txt}Observations        {res}        14225           30571           30571           29777           30571   {txt}
{txt}{hline 100}
{txt}Exponentiated coefficients; t statistics in parentheses
{txt}* p<.10, ** p<.05, *** p<.01

{com}. 
. 
. esttab using "../Results/Tables/Table_9_Panel_A.tex", replace booktabs b(%8.3f) t(%8.3f)  ///
>                 nocons nonotes starlevels(* .10 ** .05 *** .01) nolegend numbers ///
>                 eform label substitute("\_" "_") interaction(" $\times$ ") ///
>                 drop(c.ban_intensity_max_sd#c.small_firm c.ban_intensity_max_sd#c.share_low c.ban_intensity_max_sd#c.big_plant post2015 big_plant small_firm c.post2015#c.big_plant c.post2015#c.small_firm  c.post2015#c.share_low share_low) ///
>                 nomtitles prehead("{c -(} \def\sym#1{c -(}\ifmmode^{c -(}#1{c )-}\else\(^{c -(}#1{c )-}\)\fi{c )-}" ///
>                                                               "\begin{c -(}tabular{c )-}{c -(}l*{c -(}@M{c )-}{c -(}r{c )-}{c )-}" ///
>                                                               "\toprule" ///
>                                                               "& \multicolumn{c -(}5{c )-}{c -(}c{c )-}{c -(}Plant Closure{c )-} \\" ///
>                                                               "\cmidrule{c -(}2-6{c )-}" ///
>                                                               "& \multicolumn{c -(}1{c )-}{c -(}c{c )-}{c -(}Pre-Period{c )-} & \multicolumn{c -(}4{c )-}{c -(}c{c )-}{c -(}Full Sample{c )-} \\" ///
>                                                               "\cmidrule(l{c -(}2pt{c )-}){c -(}2-2{c )-} \cmidrule(l{c -(}2pt{c )-}){c -(}3-6{c )-}") ///
>                 stats(country N, fmt( 0 0 %8.0fc) labels("Country Strata" "Observations"))
{res}{txt}(output written to {browse  `"../Results/Tables/Table_9_Panel_A.tex"'})

{com}.                 
. //Panel B: Effects on Coal-fired Power Plant CO2 Emissions
. use "../Intermediate/Coal_Plant_Emissions", clear
{txt}
{com}. eststo clear
{txt}
{com}.        
. reghdfe co2_scaled treatpost2, absorb(newid i.count#i.year ) cl(year newid )  
{res}{txt}(dropped 39 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}{txt}Warning: VCV matrix was non-positive semi-definite; adjustment from Cameron, Gelbach & Miller applied.
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     3,656
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}     12{txt}){col 67}= {res}      5.96
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0311
{txt}{col 51}R-squared{col 67}= {res}    0.5590
{txt}{col 51}Adj R-squared{col 67}= {res}    0.4966
{txt}{col 1}Number of clusters ({res}year{txt}) {col 30}= {res}        13{txt}{col 51}Within R-sq.{col 67}= {res}    0.0115
{txt}{col 1}Number of clusters ({res}newid{txt}) {col 30}= {res}       330{txt}{col 51}Root MSE{col 67}= {res}    0.3344

{txt}{ralign 78:(Std. err. adjusted for {res:13} clusters in {res:year newid})}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}  co2_scaled{col 14}{c |} Coefficient{col 26}  std. err.{col 38}      t{col 46}   P>|t|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 2}treatpost2 {c |}{col 14}{res}{space 2}-.0833864{col 26}{space 2} .0341664{col 37}{space 1}   -2.44{col 46}{space 3}0.031{col 54}{space 4}-.1578286{col 67}{space 3}-.0089443
{txt}{space 7}_cons {c |}{col 14}{res}{space 2} .9440433{col 26}{space 2} .0365483{col 37}{space 1}   25.83{col 46}{space 3}0.000{col 54}{space 4} .8644114{col 67}{space 3} 1.023675
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 14}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}  Absorbed FE{col 15}{c |} Categories{col 28} - Redundant{col 40}  = Num. Coefs{col 55}{c |}
{res}{col 1}{text}{hline 14}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}        newid{col 15}{c |}{space 1}      330{col 28}{space 1}      330{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}   count#year{col 15}{c |}{space 1}      123{col 28}{space 1}      123{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}{hline 14}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est1{txt} stored)

{com}. 
. reghdfe co2_scaled  treatpost2 if active == 1, absorb(newid i.count#i.year ) cl(year newid )
{res}{txt}(dropped 42 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 8 iterations)
{res}{txt}Warning: VCV matrix was non-positive semi-definite; adjustment from Cameron, Gelbach & Miller applied.
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     3,319
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}     12{txt}){col 67}= {res}      4.45
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0566
{txt}{col 51}R-squared{col 67}= {res}    0.5417
{txt}{col 51}Adj R-squared{col 67}= {res}    0.4705
{txt}{col 1}Number of clusters ({res}year{txt}) {col 30}= {res}        13{txt}{col 51}Within R-sq.{col 67}= {res}    0.0068
{txt}{col 1}Number of clusters ({res}newid{txt}) {col 30}= {res}       329{txt}{col 51}Root MSE{col 67}= {res}    0.2956

{txt}{ralign 78:(Std. err. adjusted for {res:13} clusters in {res:year newid})}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}  co2_scaled{col 14}{c |} Coefficient{col 26}  std. err.{col 38}      t{col 46}   P>|t|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 2}treatpost2 {c |}{col 14}{res}{space 2}-.0554026{col 26}{space 2} .0262659{col 37}{space 1}   -2.11{col 46}{space 3}0.057{col 54}{space 4}-.1126312{col 67}{space 3} .0018259
{txt}{space 7}_cons {c |}{col 14}{res}{space 2} .9927635{col 26}{space 2}  .025209{col 37}{space 1}   39.38{col 46}{space 3}0.000{col 54}{space 4} .9378377{col 67}{space 3} 1.047689
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 14}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}  Absorbed FE{col 15}{c |} Categories{col 28} - Redundant{col 40}  = Num. Coefs{col 55}{c |}
{res}{col 1}{text}{hline 14}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}        newid{col 15}{c |}{space 1}      329{col 28}{space 1}      329{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}   count#year{col 15}{c |}{space 1}      117{col 28}{space 1}      117{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}{hline 14}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est2{txt} stored)

{com}. 
. reghdfe active treatpost2, absorb(newid i.count#i.year ) cl(newid)
{res}{txt}(dropped 39 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 6 iterations)
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     3,719
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}    334{txt}){col 67}= {res}      7.89
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.0053
{txt}{col 51}R-squared{col 67}= {res}    0.4853
{txt}{col 51}Adj R-squared{col 67}= {res}    0.4130
{txt}{col 51}Within R-sq.{col 67}= {res}    0.0077
{txt}{col 1}Number of clusters ({res}newid{txt}) {col 30}= {res}       335{txt}{col 51}Root MSE{col 67}= {res}    0.2179

{txt}{ralign 78:(Std. err. adjusted for {res:335} clusters in {res:newid})}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}      active{col 14}{c |} Coefficient{col 26}  std. err.{col 38}      t{col 46}   P>|t|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 2}treatpost2 {c |}{col 14}{res}{space 2}-.0443866{col 26}{space 2} .0157982{col 37}{space 1}   -2.81{col 46}{space 3}0.005{col 54}{space 4} -.075463{col 67}{space 3}-.0133101
{txt}{space 7}_cons {c |}{col 14}{res}{space 2} .9589142{col 26}{space 2} .0169589{col 37}{space 1}   56.54{col 46}{space 3}0.000{col 54}{space 4} .9255545{col 67}{space 3}  .992274
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 14}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}  Absorbed FE{col 15}{c |} Categories{col 28} - Redundant{col 40}  = Num. Coefs{col 55}{c |}
{res}{col 1}{text}{hline 14}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}        newid{col 15}{c |}{space 1}      335{col 28}{space 1}      335{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}   count#year{col 15}{c |}{space 1}      123{col 28}{space 1}        0{col 40}{result}{space 1}      123{col 54}{text} {col 55}{c |}
{res}{col 1}{text}{hline 14}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est3{txt} stored)

{com}. 
. reghdfe co2_intensity_win treatpost2, absorb(newid i.count#i.year ) cl(newid year) 
{res}{txt}(dropped 14 {browse "http://scorreia.com/research/singletons.pdf":singleton observations})
{res}{txt}({browse "http://scorreia.com/research/hdfe.pdf":MWFE estimator} converged in 7 iterations)
{res}{txt}Warning: VCV matrix was non-positive semi-definite; adjustment from Cameron, Gelbach & Miller applied.
{res}
{txt}HDFE Linear regression{col 51}Number of obs{col 67}= {res}     1,986
{txt}Absorbing 2 HDFE groups{col 51}F({res}   1{txt},{res}     10{txt}){col 67}= {res}      0.04
{txt}Statistics robust to heteroskedasticity{col 51}Prob > F{col 67}= {res}    0.8395
{txt}{col 51}R-squared{col 67}= {res}    0.7683
{txt}{col 51}Adj R-squared{col 67}= {res}    0.7263
{txt}{col 1}Number of clusters ({res}newid{txt}) {col 30}= {res}       245{txt}{col 51}Within R-sq.{col 67}= {res}    0.0000
{txt}{col 1}Number of clusters ({res}year{txt}) {col 30}= {res}        11{txt}{col 51}Root MSE{col 67}= {res}    0.2153

{txt}{ralign 78:(Std. err. adjusted for {res:11} clusters in {res:newid year})}
{hline 13}{c TT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{col 14}{c |}{col 26}    Robust
{col 1}co2_intens~n{col 14}{c |} Coefficient{col 26}  std. err.{col 38}      t{col 46}   P>|t|{col 54}     [95% con{col 67}f. interval]
{hline 13}{c +}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{space 2}treatpost2 {c |}{col 14}{res}{space 2}-.0040313{col 26}{space 2} .0193888{col 37}{space 1}   -0.21{col 46}{space 3}0.839{col 54}{space 4}-.0472322{col 67}{space 3} .0391696
{txt}{space 7}_cons {c |}{col 14}{res}{space 2} 1.061464{col 26}{space 2} .0230279{col 37}{space 1}   46.09{col 46}{space 3}0.000{col 54}{space 4} 1.010155{col 67}{space 3} 1.112774
{txt}{hline 13}{c BT}{hline 11}{hline 11}{hline 9}{hline 8}{hline 13}{hline 12}
{res}
{txt}Absorbed degrees of freedom:
{res}{col 1}{text}{hline 14}{c TT}{hline 12}{hline 12}{hline 14}{hline 1}{c TRC}
{col 1}{text}  Absorbed FE{col 15}{c |} Categories{col 28} - Redundant{col 40}  = Num. Coefs{col 55}{c |}
{res}{col 1}{text}{hline 14}{c +}{hline 12}{hline 12}{hline 14}{hline 1}{c RT}
{col 1}{text}        newid{col 15}{c |}{space 1}      245{col 28}{space 1}      245{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}   count#year{col 15}{c |}{space 1}       59{col 28}{space 1}       59{col 40}{result}{space 1}        0{col 54}{text}*{col 55}{c |}
{res}{col 1}{text}{hline 14}{c BT}{hline 12}{hline 12}{hline 14}{hline 1}{c BRC}
* = FE nested within cluster; treated as redundant for DoF computation
{res}{txt}
{com}. FE1

{txt}added macro:
      e(borrower_year) : "{res:Yes}"

added macro:
               e(bank) : "{res:Yes}"

{com}. eststo
{txt}({res}est4{txt} stored)

{com}. 
. 
. 
. esttab, label starlevels(* .10 ** .05 *** .01) nocons
{res}
{txt}{hline 84}
{txt}                              (1)             (2)             (3)             (4)   
{txt}                       co2_scaled      co2_scaled          active       co2_i..01   
{txt}{hline 84}
{txt}$\text{Bank Exit E~}{res}      -0.0834**       -0.0554*        -0.0444***     -0.00403   {txt}
                    {res} {ralign 12:{txt:(}-2.44{txt:)}}    {ralign 12:{txt:(}-2.11{txt:)}}    {ralign 12:{txt:(}-2.81{txt:)}}    {ralign 12:{txt:(}-0.21{txt:)}}   {txt}
{txt}{hline 84}
{txt}Observations        {res}         3656            3319            3719            1986   {txt}
{txt}{hline 84}
{txt}t statistics in parentheses
{txt}* p<.10, ** p<.05, *** p<.01

{com}. 
. esttab using "../Results/Tables/Table_9_Panel_B.tex", replace booktabs b(%8.3f) se(%8.3f)  ///
>                 nocons nonotes starlevels(* .10 ** .05 *** .01) nolegend numbers ///
>                 label interaction(" $\times$ ") substitute("\_" "_") ///
>                 nomtitles prehead("{c -(} \def\sym#1{c -(}\ifmmode^{c -(}#1{c )-}\else\(^{c -(}#1{c )-}\)\fi{c )-}" ///
>                                                               "\begin{c -(}tabular{c )-}{c -(}l*{c -(}4{c )-}{c -(}r{c )-}{c )-}" ///
>                                                               "\toprule" ///
>                                                               "& \multicolumn{c -(}1{c )-}{c -(}c{c )-}{c -(}Emissions{c )-} & \multicolumn{c -(}1{c )-}{c -(}c{c )-}{c -(}Active Facilities Only{c )-} & \multicolumn{c -(}1{c )-}{c -(}c{c )-}{c -(}Active (1/0){c )-} & \multicolumn{c -(}1{c )-}{c -(}c{c )-}{c -(}Carbon Intensity{c )-}\\" ///
>                                                               "\cmidrule{c -(}2-5{c )-}" ) ///
>                 stats(borrower cy N r2_a , fmt( 0 0 %8.0fc 3) labels("Facility FE" "Country x Year FE" "Observations" "Adj-R$^2$" ))
{res}{txt}(output written to {browse  `"../Results/Tables/Table_9_Panel_B.tex"'})

{com}. 
{txt}end of do-file

{com}. log close
      {txt}name:  {res}<unnamed>
       {txt}log:  {res}/Users/jblume/Documents/GitHub/coal_finance/submissions/JFE/Accepted_Paper_Submission/Code_Replication_Package/Log_Book/Main_Tables.smcl
  {txt}log type:  {res}smcl
 {txt}closed on:  {res}12 May 2025, 15:05:40
{txt}{.-}
{smcl}
{txt}{sf}{ul off}